Citation
Near-Surface Structural Examination of Human Tooth Enamel Subject to In Vitro Demineralization and Remineralization

Material Information

Title:
Near-Surface Structural Examination of Human Tooth Enamel Subject to In Vitro Demineralization and Remineralization
Creator:
Gaines, Carmen
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (144 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Materials Science and Engineering
Committee Chair:
Craciun, Valentin
Committee Members:
Anusavice, Kenneth J.
Sinnott, Susan B.
Dempere, Luisa A.
Takano, Yasumasa
Graduation Date:
8/9/2008

Subjects

Subjects / Keywords:
Apatites ( jstor )
Crystallinity ( jstor )
Crystallites ( jstor )
Lesions ( jstor )
Minerals ( jstor )
Nanoindentation ( jstor )
Paste ( jstor )
Surface layers ( jstor )
Teeth ( jstor )
Tooth enamel ( jstor )
Materials Science and Engineering -- Dissertations, Academic -- UF
crystallinity, demineralization, enamel, grazing, incidence, nanoindentation, remineralization, surface, tooth, xrd
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Materials Science and Engineering thesis, Ph.D.

Notes

Abstract:
The early stages of chemical tooth decay are governed by dynamic processes of demineralization and remineralization of dental enamel that initiates along the surface of the tooth. Conventional diagnostic techniques lack the spatial resolution required to analyze near-surface structural changes in enamel at the submicron level. In this study, slabs of highly-polished, decay-free human enamel were subjected to 0.12M EDTA and buffered lactic acid demineralizing agents and MI Paste and calcifying (0.1 ppm F) remineralizing treatments in vitro. Grazing incidence x-ray diffraction (GIXD), a technique typically used for thin film analysis, provided depth profiles of crystallinity changes in surface enamel with a resolution better than 100 nm. In conjunction with nanoindentation, a technique gaining acceptance as a means of examining the mechanical properties of sound enamel, these results were corroborated with well-established microscopy and Raman techniques to assess the nanohardness, morphologies and chemical nature of treated enamel. Interestingly, the average crystallite size of surface enamel along its c-axis dimension increased by nearly 40% after a 60 min EDTA treatment as detected by GIXD. This result was in direct contrast to the obvious surface degradation observed by microscopic and confocal Raman imaging. A decrease in nanohardness from 4.86 ? 0.44 GPa to 0.28 ? 0.10 GPa was observed. Collective results suggest that mineral dissolution characteristics evident on the micron scale may not be fully translated to the nanoscale in assessing the integrity of chemically-modified tooth enamel. While an intuitive decrease in enamel crystallinity was observed with buffered lactic acid-treated samples, demineralization was too slow to adequately quantify the enamel property changes seen. MI Paste treatment of EDTA-demineralized enamel showed preferential growth along the a-axis direction. Calcifying solution treatments of both demineralized sample types appeared to have negligible effects on enamel crystallinity. Both remineralizing agents provided an increase in resiliency within the enamel surface layers. Findings from this study may prove useful in identifying more effective methods to prevent enamel demineralization and to promote and/or enhance remineralization for the treatment of tooth decay. Careful consideration of the nanoscale properties of treated surface enamel may lead to an understanding of how to truly regenerate decomposed enamel mineral from the inside out. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2008.
Local:
Adviser: Craciun, Valentin.
Statement of Responsibility:
by Carmen Gaines.

Record Information

Source Institution:
UFRGP
Rights Management:
Copyright Gaines, Carmen. 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.
Classification:
LD1780 2008 ( lcc )

Downloads

This item has the following downloads:


Full Text





through abrasion. Conditions for the profilometric study of demineralized enamel are a flat,

baseline surface with reflective properties. In examining tooth enamel, the latter requires a shiny,

metallic coating. Although profilometry has proven to be a quick, reliable method to measure

mineral surface topographical changes, it also only gives an indirect measurement of near-

surface structural changes.

Mechanical Testing

Microhardness testing has long been an acceptable means of assessing mechanical property

changes upon enamel demineralization.[291 Vickers or Knoop diamond indenters are applied

under specified load and time conditions to a tooth sample. The size of the indent is measured. In

surface microhardness testing, the indenter is applied perpendicular to the flat, polished, outer

surface of the tooth. The indent size alone can give a relative indication of the extent of

demineralization, but must be correlated with a quantitative technique (typically MR) to

accurately assess mineral loss.[25] In cross-sectional microhardness, a Knoop indenter is typically

applied to a sample cut parallel to the tooth's biting surface. A mineral profile of enamel from

the outer surface of the sample inward can be obtained. This provides an indirect method to

quantitatively measure changes in mineral content. Although providing more accurate mineral

content data compared with surface microhardness, a major drawback is that the outer 25 .im of

the sample cannot be measured.[25] Mechanical changes in surface enamel due to early

demineralization cannot be accurately determined.

Although microhardness testing has been considered the standard in determining the

mechanical properties of teeth, such a relatively large-scale method cannot resolve surface

characteristics instrumental in early enamel dissolution. A more precise measurement technique

is desirable. Nanoindentation is typically used to characterize the mechanical properties of thin

films on the submicron scale. Under normal testing conditions, the material is subjected to a





























Figure 5- 1. AFM images highlighting transformation of as-received enamel upon 0.12M EDTA
demineralization for 30 min. Scale = 10 im2


Figure 5- 2. FE-SEM image highlighting transformation of as-received enamel upon 0.12M
EDTA demineralization for 30 min.









As stated earlier, enamel dissolution through acidic by-products of bacterial origin may

give rise to caries lesions. Covering an area that is 20 to 50 [im thick, they are found beneath a

well-mineralized, "protective" surface layer of highly-fluoridated enamel. The lesions are

described as partially-demineralized enamel, typically experiencing mineral loss of

30 to 50%.[12' 14] Early caries lesions appear to the naked eye as opaque 'white spots' on the

enamel surface. Microstructurally, the decay produce pitted surfaces. Caries lesions develop over

a relatively long period of time, often years. The acid-buffering effects of saliva, removal of

plaque from the tooth surface and regulating dietary sugar intake limits the rate of lesion

formation. Caries lesions serve as precursors to cavities when the decay progresses beyond the

point of repair. Conditions encouraging the possible regeneration of decayed enamel are

discussed in Section 2.4.

Dental erosion is governed by similar solubility laws followed by caries lesion formation,

as acid comes into direct contact with the enamel surface. Eroded lesions are formed by acids of

nonbacterial origin, such as those typically found in food and drink. Lesions commonly appear

on tooth surfaces that do not come in direct contact with plaque. Early erosion, also known as

enamel softening, extends to depths of 1-5 im.[16-19] Advanced-stage erosion is characterized by

the complete layer-by-layer dissolution of enamel, which can make it difficult to measure in

terms of relative mineral loss (Figure 2-7).[20] The tooth displays a polished, shiny appearance to

the naked eye, making the condition difficult to detect (Figure 2-8). Eroded lesion formation

occurs relatively quickly; it can occur within a few minutes. Enamel (and dentin)

demineralization can also be initiated by dental clinicians as a means to treat severe decay

problems through an acid-etching process. In enamel, selective crystallite dissolution occurs,

leaving behind a relatively porous surface. The tooth appears white and chalky to the naked eye,









6- 10. Representative depth profile of the average enamel crystallite size along the c-axis as
a function of co for as-received enamel treated with EDTA for 60 min, ending with a
5 m in M I PasteTM treatm ent. ................................................. ............................... 120

6- 11. Comparison of EDTA-treated and subsequently remineralized enamel nanohardness
v alu es ......................................................... .................................. 12 1

6- 12. Comparison of EDTA-treated and subsequently remineralized enamel elastic modulus
v alu es.......................................................... .................................. 122

6- 13. Comparison of representative force (load)-displacement curves for as-received,
60 min EDTA-treated and remineralized enamel .................................... ..................... 122

6- 14. X-Y surface images of 60 min EDTA + MI PasteTM-treated (5 min) enamel..................123

6- 15. Confocal microscope X-Z cross-sectional image of 60 min EDTA + MI PasteTM-
treated (5 min) enamel. ......................................... ........ .............. .. 123

6- 16. Averaged Raman spectrum of 60 min EDTA + MI PasteTM-treated (5 min) enamel. ......124

6- 17. AFM images of buffered lactic acid-treated enamel (room temperature) subjected to
MI PasteTM for 5 min ................... .......................... .......... 124

6- 18. Representative depth profile of the average enamel crystallite size along the c-axis as
a function of co for as-received and buffered lactic acid-treated enamel, ending with a
5 m in M I PasteTM treatm ent. ................................................ ................................ 125

6- 19. AFM images of 60 min EDTA-treated enamel subjected to a calcifying solution for
8 h ........................................................... ................................ . 1 2 5

6- 20. GIXD spectra comparison of as-received, EDTA- and calcifying solution-treated
enam el at co = 0.40.......................................................................126

6- 21. Representative depth profile of the average enamel crystallite size along the c-axis as
a function of co for as-received and 60 min EDTA-treated enamel, ending with a
5 min MI PasteTM and a 8 h calcifying solution treatment. ...........................................126

6- 22. AFM images of buffered lactic acid-treated enamel (room temperature) subjected to a
calcifying solution for 8 h. ....................................................................... ................... 127

6- 23. FE-SEM image of 30 min buffered lactic acid-treated enamel (room temperature)
subjected to a calcifying solution for 8 h, highlighting the central region of a single
e n a m e l ro d ..................................................................................... 12 7

6- 24. Representative depth profile of the average enamel crystallite size along the c-axis as
a function of co for as-received and 30 min buffered lactic acid-treated enamel,
ending with an 8 h calcifying solution treatm ent ............................ ...........................128


























W -' rn V % IL LA


B


Figure 4- 10. Confocal microscopy images of as-received enamel. A) X-Y surface scan. B) X-Z
cross-sectional scan.


C-H tretching band
2890


2890

/


I II


"-4, i : ;:
**1 tiEHI


-p


.LO I'll- : -k


Figure 4- 11. Averaged Raman spectrum of as-received enamel. Both apatitic and organic
components are shown.









CHAPTER 1
INTRODUCTION

Biological apatite is a major constituent of a variety of naturally-occurring hard tissues,

namely bone and teeth. Apatite dissolution is known to compromise the structural integrity and,

consequently, the desired functionality of these materials. This research examines the

degradation of the hardest tissue in the human body-tooth enamel-as it relates to the process

of decay, or demineralization.

Prevalence of Chemical Tooth Decay

Enamel demineralization, and subsequent increase in surface porosity, is characteristic of

the chemical decay and erosion often experienced by the tooth. The formation of acid in dental

plaque through the metabolism of sugary substances in vivo slowly decomposes the enamel

surface of the tooth. Left untreated, tooth decay eventually spreads to underlying layers of dental

tissue, forming highly-demineralized zones called caries lesions. A likely consequence of

advanced tooth decay is irreversible cavitation. In extreme cases, necrosis of the innervated

pulpal tissue will occur. The inconvenient, costly, and often painful, circumstances surrounding

tooth decay may require treatments ranging from protective sealant applications to root canal

surgeries. Considered the most commonly treated human oral disease to date, the battle with

tooth decay is likely to be fought throughout an individual's lifetime. According to the U.S.

Department of Health and Human Services, nearly 20% of small children aged 2-4 years have

shown signs of potential cavitation through the development of precursory caries lesions.[1] A

vast 80% of the American population has had at least one cavity by late adolescence. Over 66%

of adults aged 35-44 years have lost one permanent tooth through caries-induced decay, and

approximately 25% of adults between the ages of 65 and 74 have lost all of their permanent

teeth.[1]











Amorphous carbon

uJ/


v&I 4I~~


Figure 5- 16.


Averaged Raman spectrum of as-received enamel treated with 0.12M EDTA for
60 min. Note the prominent amorphous carbon bands due to laser burning of the
sample.


Figure 5- 17. AFM images of as-received enamel treated with buffered lactic acid for 30 min at
room temperature. Scale = 20 itm2


_ i _j III_ ~1









parameter defines the angle between the sample and incident beam. The rotation parameter, phi

(p), turns the longitudinal plane of the sample. The tilt parameter, psi (xu), allows for horizontal

tilting of the sample. The latter two parameters are controlled by the movement of the sample

stage, which has a 0-900 range of motion (Figure 3-4). The x-ray source is a tungsten filament,

which interacts with a copper anode for x-ray production.

Primary and secondary optics are placed in the line of the beam for conditioning purposes.

The primary (incident) beam optics prepares the x-ray beam for sample interaction as it leaves

the filament tube. The x-ray mirror device suppresses the Kp radiation to produce a beam

consisting primarily of K, radiation. Reflection of the beam onto a parabolic mirror converts it

into a quasi-parallel state. An automatic attenuator is attached to the x-ray mirror device to

prohibit damage to the x-ray detectors; it can be set to allow no more than 500,000 counts per

second (cps) at its maximum. After interactions with the sample, the beam is further conditioned

through the secondary optics. A parallel plate collimator, used to further decrease x-ray beam

intensity, attached to a xenon detector completes this set-up. A beam mask may be inserted into

the secondary optics set-up to decrease the height of the beam, if necessary.

X-ray reflectivity (XRR)

Specular XRR relies on the constructive interference of x-rays reflected at layered

interfaces within a sample, distinguishable by varying material electron densities. While co

remains below a critical angle 0c, specular reflection of the x-ray beam from the sample occurs.

As co increases to values above 0c, x-ray penetration occurs. The critical angle for total x-ray

reflection is sample-specific and has a value of 0.05-1.500 for most materials.[34] This value is

dependent upon the sample electron density and x-ray wavelength. Determination of 0o is based

on the refractive properties of the sample. The refractive index (n) is defined as

n = 1 6 (3-1)









drawback is a further decrease in peak intensity of the GIXD spectrum. This is a result of the

shallow x-ray attenuation length used in GIXD. Fewer crystallites are available for scattering,

therefore producing peaks of lower intensity.[36] The XRR curve will serve as a baseline for

examining shifts in 0o after selected enamel surface treatments. This action is indirectly related to

changes in enamel density as seen in Equation 3-3.

Nanohardness and elastic modulus values obtained for both as-received enamel and dentin

were comparable to values reported in the literature (Table 2-2). Acceptable data for evaluating

mechanical properties at the nanoscale using this technique included deviations from the average

calculated values of no more than + 10%. Fewer data were deemed acceptable in the analyses for

dentin as opposed to enamel. This may be due to the relatively pliable surface observed for

dentin. 118] Lateral variance in mechanical properties may be partly responsible for the error

spread seen when applying nanoindentation studies to biological materials in general.

Confocal Raman microscopy further verified the preferential orientation and structure of

tooth enamel used for this research. Raman laser scattering induced strong phosphate band

excitation in enamel, a group of which are better suited for Raman enamel studies due to their

high intensities relative to Ca-OH bands in the 300/cm-range.1119] The presence of sharp

phosphate bands, especially at 959/cm, is an indication of relatively strong, intact enamel.

Due to the anticipated high variability of enamel material properties and the precision with

which the enamel samples were characterized, using the same tooth for each complete cycle of

treatments and/or set of characterizations is desirable.

Characterization Concerns

Biological materials introduce unique challenges in experimental execution. Tooth-to-tooth

variability among individuals within a sample population can be expected based on genetics and









5- 24. Comparison of as-received and buffered lactic acid-treated enamel nanohardness
v alu es ......................................................... .................................. 10 1

5- 25. Comparison of as-received and EDTA-treated enamel elastic modulus values.............. 101

5- 26. Comparison of representative force (load)-displacement curves for as-received and
30 min buffered lactic acid-treated enam el.................................. ........................ 102

5- 27. Confocal microscopy X-Y surface image of as-received enamel treated with buffered
lactic acid for 30 m in at room temperature ..................................... ......... ............... 102

5- 28. Averaged Raman spectrum of enamel treated with buffered lactic acid for 30 min
at room tem perature.. ........................ ...... ..................... .... ............... 103

5- 29. TEM image showing the bonding of neighboring crystallites within a caries lesion........103

6- 1. AFM images of EDTA-treated enamel (30 min) subjected to MI PasteTM for 5 min ........116

6- 2. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI PasteTM for 5 min .....116

6- 3. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI PasteTM for 5 min,
highlighting a single enam el rod .......................... .......... .................... ............... 117

6- 4. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI PasteTM for 5 min,
highlighting central region of a single enamel rod. ............. ........................................117

6- 5. Representative GIXD spectra comparison of as-received, EDTA- and MI PasteTM-
treated enam el at co = 0.40 ............................................................. .. .. .............. 118

6- 6. (002) GIXD spectra comparison for as-received, EDTA-treated (30 min) and
MI PasteTM-treated (5 and 10 min) enamel at co = 0.40. .................... .........................118

6- 7. Depth profile of the average enamel crystallite size along the c-axis as a function of o
for as-received enamel progressively treated with EDTA, ending with a 5 min
M I PasteTM treatment ................................ ........... .... ...... ....... ..... 119

6- 8. Depth profile of the average surface enamel crystallite sizes along the c-axis as a
function of co for as-received enamel progressively treated with EDTA, ending with a
5 m in M I PasteTM treatm ent .................................. ............................... ............... 119

6- 9. (002) GIXD spectra comparison for as-received, EDTA-treated (30 min) and
MI PasteTM-treated (5 and 10 min) enamel at co = 0.4. ........... ...............................120











2000 -

1750 -

1500


S1250loo -

b1000-

750-

500 -

250 -

0


20 22.5


25 27.5


30 32.5


42Theta


Figure 6- 5. Representative GIXD spectra comparison of as-received, EDTA- and MI PasteTM-
treated enamel at co = 0.40. A) As-received. B) 0.12M EDTA treatment for 30 min. C)
EDTA + MI PasteTM treatment for 5 min.


600
As-Received
**.... 30 min EDTA
500
EDTA+ MI Paste (5 min)
-EDTA + MI Paste (10 min)
400


S300


200


100


0 ---------------------
25 25.5 26 26.5
02Theta



Figure 6- 6. (002) GIXD spectra comparison for as-received, EDTA-treated (30 min) and
MI PasteTM-treated (5 and 10 min) enamel at co = 0.40.


S-- As-Received
a .... 30SinEDTA
-- -EDTA+ MI Pasre (5 min)












9 '- -- --**9 ., -..






























Wavenumbers (cm-1)


Figure 5- 28. Averaged Raman spectrum of enamel treated with buffered lactic acid for 30 min
at room temperature. Note absence of organic material bands within 2000/cm to
2500/cm range.


Figure 5- 29. TEM image showing the bonding of neighboring crystallites within a caries lesion.
The white arrow indicates the central dark line. As the crystal grows toward the left, it
begins to "fuse" (black arrow) with a neighboring crystal. Lattices of the original
crystal (broken line) are 10 out of alignment with lattices of the neighboring crystal
(solid line).[39]









with the ability to conduct elemental analyses at the enamel surface, would provide a better

indication of structural mechanisms behind enamel remineralization.









near the periphery of the enamel rod walls after 60 min. Confocal Raman imaging and spectral

changes after a 60-min EDTA treatment confirmed major enamel demineralization. On the

crystalline level, GIXD detected an increase in crystallinity upon 0.12M EDTA treatment most

evident to approximately 1000 nm below the enamel surface. This counterintuitive result may be

a function of the dissolution of smaller enamel crystallites upon demineralization, or the bonded

of neighboring crystallites as a result of surface diffusion processes. Nanoindentation studies

showed appreciable difference in hardness and elastic modulus after an extensive 60-min EDTA

treatment, with values shifting from 4.86 0.44 GPa for as-received enamel to 0.28 + 0.10 GPa

and from 102.63 18.87 GPa to 4.82 1.25 GPa, respectively. Although errors may convolute

the true effects of the treatments on the mechanical properties calculated, these relative changes

appear reasonable. Collective results suggest that mineral dissolution evident on the micron scale

may be governed by an increase in enamel crystallinity on the nanoscale, leading to extensive

decomposition and highly increased fragility of tooth enamel. While an intuitive decrease in

enamel crystallinity was observed with buffered lactic acid-treated samples, demineralization

was too slow to adequately quantify the enamel property changes seen.

The remineralization of extensively-demineralized enamel substrates showed preferential

growth along the (200) plane crystallitee a-axis dimension) upon treatment with a CPP-ACP

topical paste. This did not translate to an increase in nanohardness of the enamel, as its average

value remained relatively constant at 0.2 0.2 GPa. The elastic modulus increased to

19.1 + 13.5 GPa, suggesting an increase in mechanical resiliency upon remineralization.

Structural effects of the calcifying solution (0.5 ppm F) were inconclusive as a surface layer

appeared to form on the enamel substrate. While remineralization studies with the topical paste

and calcifying solution were also conducted on buffered lactic acid-demineralized enamel









Confocal R am an M icroscopy .............................................. ....... ........................ 54
Scanning Electron M icroscopy (SEM )................................ ........................ ........ 54
Atom ic Force M icroscopy (AFM )......................................................... ............... 55
Attempted TEM Characterization ...........................................................................55
Statistical C considerations ...................... .................. ................. ... ....... ..56

4 EN AM EL CH A RA CTER IZA TION ............................................................. ....................62

In tro du ctio n ................... ...................6...................2..........
M materials an d M eth o d s ...................................................................................................... 6 3
A F M .............. ... ................................................................63
S E M ................... ...................6...................3..........
X R D ................... ...................6...................4..........
N anoindentation .................................................................. 64
Confocal Ram an M icroscopy .............. .............................................. ...............65
R e su lts ................................................................................................................................ 6 5
D isc u ssio n .................................. .........................................................6 6
T ooth E nam el D defined ..............................................................66
C haracterization C concern s.......................................................................................... 67

5 ENAMEL DEMINERALIZATION ................................................................................ 75

In tro du ctio n ................... ...................7...................5..........
M materials and M ethods ....................................................................................... ............... 77
Enam el D em ineralization Treatm ents ....................................................... 77
Demineralized Enamel Characterization .......................................... 77
Results ................... ............................................ .......... ........ 78
Demineralization 0. 12M EDTA ................................. ......................... ...78
Demineralization Buffered Lactic Acid ..........................................81
D isc u ssio n .............. ..... .......... ......................................................................................... 8 2

6 ENAMEL REMINERALIZATION ................. ........... ............ 104

Introduction ............. ........................................104
M materials and M ethods .............................................................106
Enamel Remineralization Treatments ........................................ ... ........ 106
Remineralized Enamel Characterization ....................... .. ........................ 106
R e su lts ............. ..... .. .. .............. .. ........................................................... 10 7
M I P asteTM T reatm ents ................... ....... ......... .. ............................... ............... 107
EDTA-treated enamel substrate ................................................................. 107
Buffered lactic acid-treated enamel substrate ................................... ....109
Calcifying Solution Treatm ents................................................................................ 109
ED TA -treated enam el substrate ........................................................................ 109
Buffered lactic acid-treated enam el substrate ........................................................ 110
D iscu ssion ...................................... ................ .......... 110
M I PasteTM Treatm ents ..................................................... ............... ............... 110
Calcifying Solution Treatments ................................. ........ ...... ...............114


6









CHAPTER 3
EXPERIMENTAL METHODS

Sample preparation of the dental enamel used in this experimental study will be discussed

in this chapter. The rationale behind the selection of demineralization and remineralization

agents is explained. Finally, the characterization techniques used to identify structural,

morphological, mechanical and chemical changes in treated enamel are highlighted.

Enamel Sample Preparation

Extracted human incisor teeth were collected from the University of Florida dental clinics

for use in this research. The dental history of the tooth donors was unknown. The teeth were

sterilized by gamma radiation, using a Co60 radiation source at 0.5 Mrad for at least 9 hours.[72]

After sterilization, the teeth were stored in 0.1% thymol until needed.

The sterilized teeth were set in a cold-cure epoxy (EpofixTM, Cleveland, OH) for ease of

handling during sample preparation. First, the extraneous root sections of the teeth were ground

away using 60 grit SiC abrasive paper. The teeth were positioned labial-side down to allow for

easier access to the enamel surface. The epoxied samples were then successively ground with

SiC abrasive papers to a 1200 grit finish. Alumina slurries, ending with a 0.05 tm grit

suspension, provided a highly-polished, flat enamel surface of approximately 5.0 mm2. Care was

taken to remove at least 100 .im of enamel, eliminating any traces of FAP for standardization

purposes (Appendix A).[73] Finally, the samples were sliced into 2.0 + 1.0 mm thick slabs with a

low-speed diamond saw. These highly-polished enamel samples were termed "as-received," as

they had not yet been subjected to any chemical surface treatments (Figure 3-1). The teeth were

then ultrasonically cleaned with distilled water and stored in a hydrated environment, awaiting

characterization or further treatment.













Pmax


I
I
I
I
I
I
I
I
II
I
I
I
I


hf hmaX
Displacement (h)


Figure 2- 12. Typical load-displacement curve produced by nanoindentation.[68]








Table 2- 2. Sampling of sound tooth enamel hardness and modulus values.
Lead Author Hardness (GPa) Elastic Modulus (GPa)
Bhushan, B. (Ed.) [69] 2.7 5.2 89 -106
Finke, M [18 3.51 0.90 102.45 18.39
Habelitz, S [32] 3.9 0.3 87.5 2.2
Barbour, ME [31] 4.74 + 0.14 104.8 + 2.8
Lippert, F [33 4.55 0.17 106.0 + 2.8











2110-


200:'
02-02 07_U 043.1Tys
0420 _Lu3 O043.hys
I- 1 '"' ," i. o."


eI


BOO
.,, -- -- .-,- -- -

A / (






.06 -. --


oo0 2;O 50D 13o 102,0 12;iO liOO 17U 20V0 0 225,0 2Ou ZMO 3000 WO,0 350Q0 i 6
Digplacnaan (nnj


Figure 5- 26. Comparison of representative force (load)-displacement curves for as-received and
30 min buffered lactic acid-treated enamel. A) As-received. B) Human body
temperature treatment. C) Room temperature treatment.


Figure 5- 27. Confocal microscopy X-Y surface image of as-received enamel treated with
buffered lactic acid for 30 min at room temperature.


x TzI+i









5- 7. Depth profile of the average enamel crystallite size along the c-axis as a function of o
for ED TA enam el treatm ent regim e. ............................................................................ 92

5- 8. Depth profile of the average surface enamel crystallite sizes along the c-axis as a
function of co for EDTA enamel treatment regime .................................... .................92

5- 9. Correlation of EDTA treatment time with crystallite size at co = 0.150. .............................93

5- 10. Representative XRR curve comparison of as-received and EDTA-treated (30 min)
e n a m e l ............................................................................................ 9 3

5- 11. Comparison of as-received and EDTA-treated enamel nanohardness values...................94

5- 12. Comparison of as-received and EDTA-treated enamel elastic modulus values..................95

5- 13. Comparison of representative force (load)-displacement curves for as-received and
EDTA-treated enamel. .................................. .. .. ........ .. ............. 95

5- 14. Comparison of representative force (load)-displacement curves for EDTA-treated
enam el and as-received dentin............................................... ..................................... 96

5- 15. Confocal microscopy images of as-received enamel treated with 0.12M EDTA for
60 m in ........................................................................................... 96

5- 16. Averaged Raman spectrum of as-received enamel treated with 0.12M EDTA for
60 m in .......... ......................................... ..... 97

5- 17. AFM images of as-received enamel treated with buffered lactic acid for 30 min at
room tem perature ............... ........................... ............ ....................... 97

5- 18. AFM images of as-received enamel treated with buffered lactic acid for 30 min at
hum an body tem perature.......................................................... ....................................98

5- 19. SEM images of as-received enamel treated with buffered lactic acid for 30 min at
room tem perature ...................................... ................................................ 98

5- 20. Depth profile of the average enamel crystallite size along the c-axis as a function of co
for a 30 min buffered lactic acid treatment at human body temperature.........................99

5- 21. Representative GIXD spectra comparison of as-received and buffered lactic acid-
treated enamel at co = 0.40........... ..................... .... ...................... .. .......... 99

5- 22. Depth profile of the average enamel crystallite size along the c-axis as a function of co
for a 30 min buffered lactic acid treatment at room temperature. ..................................100

5- 23. Representative XRR curve comparison of as-received enamel and treated with
buffered lactic acid for 30 min at room temperature. .................... ................... .......... 100









CHAPTER 5
ENAMEL DEMINERALIZATION

Introduction

The study of enamel dissolution in vitro is performed by the intentional acid attack of the

apatite surfaces. While conditions such as experimental temperature, acid type and attack

frequency/duration are regulated to mimic oral conditions as closely as possible in caries lesions

and fluoride efficacy testing, acid production by bacterial metabolism of sugars in vivo can only

be simulated. Other factors, such as the ever-changing concentrations of Ca, PO4, F and other

trace elements in oral fluids and the buffering effects of the acquired pellicle absorbed onto the

enamel surface, are difficult to model. A goal of this research is to examine structural changes in

surface enamel upon demineralization at the submicron level. As such, etched enamel-which

exploits the intentional acid attack of enamel for restorative purposes-appeared a suitable

means by which to examine enamel material property changes upon demineralization. For

comparison purposes, buffered lactic acid was used as an additional demineralization agent to

better simulate enamel dissolution activities in vivo.

EDTA is an etchant used to condition tooth enamel and dentin for use in dental

restorations. A highly-roughened, porous surface is produced, increasing the surface area

available for adhesion of sealants and other bonding materials to enamel. In cases of extensive

tissue damage, collagen tubule exposure in dentin is needed to enhance attachment to this softer

region, creating a suitable interfacial layer.[77] 37% phosphoric acid at pH< 2.5 is most

commonly used for etching purposes. 79, 121] A 30 s treatment is strong enough to penetrate the

enamel and permeate into dentin, effectively causing collagen dissolution. Unfortunately,

excessive acid penetration experienced at higher acid concentrations may create dissolution

channels so deep to where the restorative material cannot reach, introducing voids and potential










Table 2- 1. Minor inorganic constituents in sound human tooth enamel.
Concentration Range Sub s [63]
Constituent (wt)[62] Substitutes For
(wt%) 62


COs- 2.4-4.2 P04-
F* 0.05-0.9 OH-, C1-
Mg2+ 0.04-0.68 Ca2
Na+ 0.17-1.16 Ca2
C1 0.16-0.7 OH-
*Present in highly variable amounts; F often considered a trace element.


Effect on Enamel
Crystallinity[631
Decrease
Increase
Decrease
None
None


Qo
( =,. p




%( : ca



Figure 2- 5. The idealized crystal structure of HAP, as viewed along the c-axis61]

Figure 2- 5. The idealized crystal structure ofHAP, as viewed along the c-axis.[611


,2,















15 1 I I I I I
0.15 0.2 0.25 0.3 0.4 0.8 1.0 T-2T
Grazing Incidence Angle, co ()
Figure 4- 8. Depth profile of the average crystallite sizes along the c-axis from the enamel
surface into the bulk.


02 02-1)_ Ii 11I: Iys /



w/


Figure 4- 9. Representative force (load)-displacement nanoindentation curve for as-received
enamel.


I' '


,1-.1 Il ci i ilII ILIL I









Considering the high mineral content of enamel, evaluation of crystallinity may provide

valuable information to better understand the structural implications of early enamel dissolution

and remineralization. An important aspect is the examination of the chemical reactions involved

in these processes. Since these reactions occur in vivo at the tooth-oral fluid interface, surface

enamel should be investigated.

Enamel Demineralization

In vivo demineralization occurs through the chemical attack of tooth enamel, resulting in

the loss of calcium and phosphate ions (as well as other substitutional impurities present, such as

carbonate) through dissolution and diffusion. Under ideal conditions saliva is supersaturated with

respect to enamel components, containing natural levels of calcium and phosphate at neutral pH.

This encourages homeostasis within the mouth.[12] Eating and drinking tends to upset this

balance. In caries lesion development, oral bacteria-specifically Streptococcus mutans

(S. mutans) and lactobacilli-convert dietary sugars into weak acids (e.g. lactic, acetic). The

bacteria, sugar, and saliva combine to form a biofilm of plaque, which can adhere to tooth

surfaces within 20 min of eating.[9] If not removed on a routine basis, the acidic plaque begins to

decompose the enamel surface. Acid production causes the plaque pH to fall below the critical

value of approximately 5.5, favoring enamel dissolution (Figure 2-6). Basic solubility laws allow

for diffusion of the Ca2+, PO43-, and OH- ions from the apatite crystals, through the plaque to the

saliva, compromising the strength and structure of enamel. This process is defined by

Equation 2-1. Acid dissolution occurs because the enamel becomes undersaturated with respect

to its most

Calo(P04)6(OH)2 + 8H+ 10Ca2++ 6HP042- + 2H20 (2-1)

important structural constituents due to the removal of these critical ions.[15]









enamel-rich ions available for remineralization.[44-46, 89] MI PasteTM (CG Corporation, Tokyo,

Japan), a topical cream containing a casein phosphopeptide-amorphous calcium phosphate

(CPP-ACP) complex, provides a reservoir of freely-available calcium and phosphate ions when

adhered to the surface of the tooth. CPP has the ability to bind to amorphous calcium phosphate,

stabilizing these ions in solution.190] The bioactive ACP works to release these ions from enamel

to maintain a supersaturated state near the enamel surface.[91] Based on RecaldentTM (originators

of CPP-ACP complex) technology, MI PasteTM may serve as a bioavailable source of calcium

and phosphate without the influence of fluoride. Various studies have reported its efficacy in the

remineralization of eroded and carious subsurface lesions.[91-93]

MI PasteTM was applied to demineralized enamel slabs for up to 10 min, in 5 min intervals.

Following each treatment, the cream was removed, rinsed and sonicated for 1 min in distilled

water.

Enamel Characterization

Acids involved in enamel dissolution must first interact with and penetrate the surface of

the tooth in vivo. A mineral density gradient forms as a result, with a decrease in mineral content

in areas affected by the acid as it travels more deeply into the dentition. The structural integrity

of the demineralized enamel layers is compromised with respect to bulk enamel, causing a

discrepancy in the material properties displayed by both regions. Early enamel dissolution, which

produces 'softened' enamel, is defined as demineralization that extends 2 to 5 .im from the

physical tooth surface.[17] Remineralization of early demineralized lesions works to regenerate

lost mineral, whose effects can be fully realized within the softened enamel regime.

Dimensionally, treated enamel is on the scale of that of a thin film, ranging from an atomic layer

to several micron thicknesses. Thin films have found application in the semiconductor,

microelectronics, communications and material coating industries where the use of small-scale











100

10


1-

0.1

0.01


0.001


0.01


100


Incidence Angle ()


Figure 3- 7. X-ray attenuation length of Cu K, radiation in HAP.















Table 3-3. Raman frequencies and P043- vibrational modes associated with tooth enamel.[103]


Vibrational Mode


Freauencv (cm )


V3 1045
vi 960
V4 590
v, 430


***

4.-b.
4+


F6


I +












34

S32

30
N
< 28

S26

U 24


0.15 0.2 0.25 0.3 0.4 0.8 1.0 T-2T


Grazing Incidence Angle, w ()



Figure 6- 18. Representative depth profile of the average enamel crystallite size along the c-axis
as a function of co for as-received and buffered lactic acid-treated enamel, ending with
a 5 min MI PasteTM treatment.


Figure 6- 19. AFM images of 60 min EDTA-treated enamel subjected to a calcifying solution for
8 h. Scale = 20 [m2


-4-As-Received 30 min Room Temp

MT Paste (5 min)





_,_ _
~~.ZI7










100I -
So-



400 -



200 -
200 -


0


"2Theta

Figure 5- 5. Representative GIXD spectra comparison of as-received enamel treated with
< 25 ml of 0.12M EDTA solution at co = 0.80. A) 10 min treatment. B) 60 min
treatment.


1300

1050

S8 00


S550


25 27.5 30 32.5
02Theta


Figure 5- 6. Representative GIXD spectra comparison of as-received and EDTA-treated enamel
at co = 0.40. A) As-received. B) 0.12M EDTA treatment for 30 min. C) 0.12M EDTA
treatment for 60 min.


19 min .... 60Minm



'^ a9


B
'.. 141 1A
oA
A ^A^^^.
fS -
CI-^4^^









10-min EDTA treatment of as-received enamel (Figure 6-8). A surprising consequence of

remineralization was the enhanced crystallinity depicted by the (200) and, to a lesser degree,

(111) reflections shown in Figure 6-5. Evidence of increased (200) peak intensity/narrowing with

remineralization is highlighted in Figure 6-9, which was enhanced with a second application of

paste. In terms of the stability of remineralized enamel treated by MI PasteTM, an average

increase in enamel crystallinity in the c-axis direction of 5 nm was determined after immersion

of the remineralized enamel slab in distilled water for 24 h (Figure 6-10). The results depicted in

this figure are based on the GIXD analysis of a single enamel slab.

Results of nanoindentation testing for the MI PasteTM remineralization potential of EDTA-

treated enamel, as compared to previously-reported as-received and demineralized enamel

values, are summarized in Table 6-1. Nanohardness results are comparable to those calculated

for 60-min EDTA-treated enamel, although the former showed a slight advantage. There is

overlap in the calculated values, making it difficult to propose significant differences in enamel

hardness (Figure 6-11). The average elastic modulus value for the remineralized enamel was an

order of magnitude higher than that of the EDTA-treated enamel (Figure 6-12). Representative

load vs. displacement curves are shown in Figure 6-13.

Confocal Raman results were extracted from the region enclosed in the red square from the

video image of textured enamel shown in Part A of Figure 6-14. Confocal imaging of this area

produced an X-Y scan in Part B of the figure, highlighting patches of apatite (red) among

potential residue from the paste (blue) and a combination of both (pink). The lower region

burned by the laser was omitted as indicated. A depth profile of the sample surface layers in

Figure 6-15 showed sporadic bright red areas of apatite within the paste residue. Black regions

within the confocal images likely reflect underfocused sample surface areas. The averaged









Confocal imaging was performed on selected remineralized enamel samples to confirm

microstructural changes observed by AFM/SEM. These images, in combination with Raman

spectra extracted from color-coded enamel phases, provided information regarding the chemistry

within the enamel surface layers.

Results

MI PasteTM Treatments

EDTA-treated enamel substrate

Figure 6-1 shows AFM images of demineralized enamel (0.12M EDTA for 30 min) after a

5-min treatment with MI PasteTM. Irregularly-shaped, raised patches were dispersed uniformly

along the surface, with little evidence of the extent of demineralization observed in Figure 5-1.

An averaged RMS roughness value of 29.2 nm was calculated over the entire image. A

secondary electron FE-SEM image of a similarly-prepared sample is shown in Figure 6-2. The

enamel rod outlines are more defined, with a greater accumulation of material within the center

of the enamel rods and along their peripheries (Figure 6-3). An image focusing on the central

region of a single enamel rod shows bundled regions of what appears to be enamel crystallites

(Figure 6-4).

A representative GIXD spectra depth profile (co = 0.40) of enamel treatments, culminating

with a 5-min MI PasteTM application, showed a pronounced decrease in crystallinity along the

crystallite c-axis direction after remineralization (Figure 6-5). This is clearly depicted by the

(002) relative peak intensity distribution in Figure 6-6. A second 5-min application of the paste

did little to enhance crystallinity. No new phases, indicated by the emergence of new diffraction

peaks, were seen. A depth profile graph of crystallite size changes from the surface toward bulk

enamel in Figure 6-7 shows little evidence of improved crystallinity with remineralization in the

[002] direction. A decrease in crystallite size was detected on the scale of that observed after a

107









in the oral environment (Figure 2-16). The process of apatite remineralization has been described

by researchers in one of three ways[39]

* Regeneration of partially-dissolved crystals
* Emergence of newly-formed crystals
* Additional growth of existing crystals

Fluoride is important to the remineralization process as it renders the enamel more resistant

to acid dissolution. Fluoride absorption into the enamel has also been seen to enhance the

chemical reactions that allow for apatite redeposition. The presence of relatively small fluoride

ions in the apatite lattice reduces the free energy of the enamel surface, thereby inhibiting the

absorption of harmful salivary impurities.[12] Increased strength and resistance to fracture and

wear of the enamel is due to the formation of fluorapatite (FAP), which dominates the upper

100 .m of surface enamel.[43] The semipermeable nature of enamel allows for the absorption of

F- from saliva to form FAP, strengthening the enamel structure and protecting the tooth against

subsequent acid attack (Equation 2-4). The most effective incorporation of fluoride

Calo(P04)6(OH)2 + 2F- Calo(P04)6F2 + 20H- (2-4)

in the oral environment is through the use of topical agents, such as fluorinated toothpastes,

mouthrinses and waters.[44] These agents provide a reservoir of Ca2+ and P043- and F- available

for diffusion into surface enamel. At relatively high F concentrations, a CaF2-like layer of

material may form on the surface.[45-47] Fluoride efficacy in enamel remineralization has been

studied extensively through pH cycling studies.[48-50] The enamel sample is subjected to a series

of demineralizing and remineralizing solutions over a specified period of time. Each solution is

saturated with respect to HAP and FAP to simulate the formation of caries lesions in vitro.[51

Fluoride may be added in specified concentrations to the treatment solutions, or applied topically

during cycling, to evaluate its reported positive effects on enamel stability. This action most









substrates, it was difficult to determine significant changes seen due to the limited

demineralization effects experienced by the samples.

Although most of the results from this research cannot be substantiated through statistical

means due to limitations in samples sizes for each treatment and other resources, many collective

sets of data showed anticipated trends that may be considered in future studies of the surface

structural effects of demineralization and remineralization.

Future Work

General trends and variability in calculated enamel crystallinity, hardness and modulus

suggests that the enamel dissolution and remineralization are much more complicated processes

than originally thought. In order to substantiate the findings of this research, additional studies in

the area of surface crystallinity are needed. Studies conducted on a larger scale would help in

verifying trends in such a variable experimental substrate as tooth enamel. Dissolution studies

incorporating a constant-composition experimental design would ensure that adequate solution

ion concentration levels are maintained in order better control the demineralization process.

HR-TEM studies would provide information on the true action of enamel crystallites subject to

EDTA demineralization, namely whether crystallite bonding does truly occur. The

cross-sectional examination of enamel surface layers with the TEM may be achieved upon

improvements in sample preparation methods realized by the FIB. The brittle nature of as-

received enamel, and potential high reactivity of treated enamel with the ion beam, makes this

technique a difficult one. If sample preparation is achieved, a depth profiling of the enamel

electron density changes upon treatment can be correlated with surface x-ray scattering studies.

High-resolution EDS elemental analysis would also provide insight into possible apatite phase

changes so slight that they are difficult to discern with XRD studies alone. Finally, a refinement

of the nanoindentation testing process for treated enamel may prove useful in accounting for









the micron scale. The convergence of calculated crystallite sizes and the great overlap

experienced in the bulk suggests that the acid solutions only affected the upper surface layers in

this study. The selected composition of the lactic acid solution illicits demineralization that can

take several days or weeks to produce artificial caries lesions, and may take a similar amount of

time for discernable demineralization indicators to appear on the enamel surface. XRR curve

comparisons suggest a negligible change in mineral density upon demineralization, yet an

increase in surface roughness.

Based on the nanoindentation results observed in this study, enamel hardness and stiffness

is compromised with extent of EDTA demineralization. Extensive EDTA demineralization

yields enamel deformation closest to that of dentin in this study, although dentin appears to be

more resilient. Subsurface demineralization of the buffered lactic acid-treated samples may have

occurred, accounting for limited enamel surface roughness observed by AFM and SEM. The

increase in mechanical property values imply that demineralization has occurred, although

enamel integrity appears to be at a relatively optimal level under human body demineralization

conditions as seen in Figure 5-26. The increase in enamel crystallite size on the nanoscale does

not appear to increase the overall hardness or strength of enamel. This effect may be due to

nanoindentation probe placement and/or the inhomogeneous nature of demineralized enamel.

Erroneous data may be produced from areas where the probe cannot achieve adequate contact

with the sample surface. Sample roughness will also limit the accuracy of nanoindentation

results.[18, 33, 132] While more consistent results were achieved with as-received enamel, there is

higher data variability when greater surface irregularities are introduced. As such,

nanoindentation results of treated enamel are better discussed in relative terms.










45
-4-As-Received -- 10min l- 30 min ---60min
40

5 35

S30

25

20


15 -.....
0.15 0.2 0.25 0.3 0.4 0.8 1.0
Grazing Incidence Angle, c (0)


Figure 5- 7. Depth profile of the average enamel crystallite size along the c-axis as a function of
co for EDTA enamel treatment regime.


0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45
Grazing Incidence Angle, o (0)

Figure 5- 8. Depth profile of the average surface enamel crystallite sizes along the c-axis as a
function of co for EDTA enamel treatment regime. Error bars represent 95%
confidence.









observed in TEM studies as mentioned in Chapter 2. As dissolution continues, the crystallite

hollows, leaving a tubular shell of apatite.[129] Further dissolution causes fracturing of these

hollow tubes into mere fragments of apatite. In this study, perhaps smaller-sized enamel

crystallites that have experienced some degree of dissolution previously had completely

dissolved after treatment. When performing XRD experiments, the diminishing effects of these

smaller crystallites would not be a contributing factor in peak geometry. As such, the XRD peaks

would not exhibit the broadening effects resulting from the presence of smaller crystals. Such an

occurrence may have basis in the Gibbs-Thompson[130] effect for nanoscale particle dissolution

and has been previously reported as an effect of the acid demineralization of bone.1131] Enamel

dissolution kinetics may also play a role in the observed increase in crystallinity. Crystallite

surface diffusion occurring during the enamel dissolution process may encourage the bonding, or

"fusion," of neighboring crystallites, potentially creating larger regions of planar alignment. This

phenomenon has been observed in HR-TEM enamel caries lesion development and enamel

crystal growth studies.[39' 71] Since XRD crystallinity is detected as crystal grains of similar lattice

orientation, it would be enhanced in this case. That is, larger crystallites would be detected by

XRD. It is important to note that crystallite bonding does not lead to perfectly-aligned lattices by

the individual crystals. Defects such as edge dislocations and small boundary grain boundaries

can be expected from such a process. The literature has cited instances of lattice mismatch

between bonded crystallites on the order of 1-2, a negligible consideration for XRD detection

(Figure 5-29). XRR studies showed a decrease in density and increase in roughness upon EDTA

demineralization.

A decrease in enamel crystallite size closest to the surface in the buffered lactic acid

experiments may reflect a dissolution condition synonymous to that which has been observed on













45

S40

^ 35
N
30

S25

0 20


30 min Room Temp


0.15 0.2 0.25 0.3 0.4 0.8 1.0 T-2T

Grazing Incidence Angle, w (")


Figure 5- 22. Depth profile of the average enamel crystallite size along the c-axis as a function of
co for a 30 min buffered lactic acid treatment at room temperature.


1.0E+06


1.0E+05


1.0E+04


1.0E+03


1.0E+02


1.OE+01 I I I I
0.2 0.5 0.7 1.0 1.2 1.5 1.7 2.0

02Theta



Figure 5- 23. Representative XRR curve comparison of as-received enamel and treated with
buffered lactic acid for 30 min at room temperature.


-w----- -


--As-Received


m















































Incident Beam








.. ,


Reflected Beam


Diffracted Beam


Figure 3- 4. PANalytical X'PERT MRD PRO four-axis set-up. A) Goniometer and secondary
optics. B) Goniometer angle designations.


9















Lingual


Buccal


5.5

SXu et aL,
1998
(occlusal
H 5 section,
Vickers)
J4.( I




3 I

2.5 Willems et
al., 1993
(nano-
indentation)


Xu et al.,
1998
(occlusal
section,
modified
Vickers)


Xu et al.,
1998 (axial
section,
I modified
Vickers)


\ Xu et al.,
1998
(axial
section,
Vickers)


Buccal


120 Willems et al., 1993
(nanoindentation)

0 Xu et al., 1998
S(occlusal section,
Craig et al., modified Vickers)
1961 (cusp,
100 compression
testing)

90


80


70
7 Staines et
S1 Xu et aL,
Craig et al., a.,1981 1998 (axial
60 1961 (side, (sphericl section,
compression indentation) modified
testing) Vickers)


Figure 4- 1. Average values of enamel hardness (a) and elastic modulus (b) as reported
earlier by other researchers for the mesial half of a maxillary M2 as determined by
nanoindentation. The standard deviations for these averages range from 0.2 to 0.3
GPa for hardness and from 2 to 5 GPa for elastic modulus.[116]


I


j~7


















T-. *-l *r p^ B* Pi ftaL1i~i


d. sine ot
** 0 f& = X 2d sin 9


* r ** *


Bragg's Law


Figure 2- 13. XRD schematic illustrating Bragg's Law.[701


Figure 2- 14. TEM image of a cross-section of an enamel crystallite c-axis. Arrow points to a
central dark line.[39]


&Wdl9
poFmp? aWSV


'
.-,9~sr, r
"1
c
c
I
-
: `.

































Figure 2- 3. Photograph of a lower first premolar human tooth slice showing enamel rod
orientation. [60]






Gingiva


',1 Cementum
Nerves and
Blood Vessels


Figure 2- 4. Schematic highlighting major components of the tooth.[61]









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

NEAR-SURFACE STRUCTURAL EXAMINATION OF HUMAN TOOTH ENAMEL
SUBJECT TO IN VITRO DEMINERALIZATION AND REMINERALIZATION

By

Carmen Veronica Gaines

August 2008

Chair: Valentin Craciun
Major: Materials Science and Engineering

The early stages of chemical tooth decay are governed by dynamic processes of

demineralization and remineralization of dental enamel that initiates along the surface of the

tooth. Conventional diagnostic techniques lack the spatial resolution required to analyze near-

surface structural changes in enamel at the submicron level. In this study, slabs of highly-

polished, decay-free human enamel were subjected to 0.12M EDTA and buffered lactic acid

demineralizing agents and MI PasteTM and calcifying (0.1 ppm F) remineralizing treatments in

vitro. Grazing incidence x-ray diffraction (GIXD), a technique typically used for thin film

analysis, provided depth profiles of crystallinity changes in surface enamel with a resolution

better than 100 nm. In conjunction with nanoindentation, a technique gaining acceptance as a

means of examining the mechanical properties of sound enamel, these results were corroborated

with well-established microscopy and Raman techniques to assess the nanohardness,

morphologies and chemical nature of treated enamel.

Interestingly, the average crystallite size of surface enamel along its c-axis dimension

increased by nearly 40% after a 60 min EDTA treatment as detected by GIXD. This result was in

direct contrast to the obvious surface degradation observed by microscopic and confocal Raman

imaging. A decrease in nanohardness from 4.86 0.44 GPa to 0.28 + 0.10 GPa was observed.









sizes, lattice defects and interplanar strain. The peak intensity provides information about the

quantity of a particular phase in the material while peak width (indicated by the full width at half

maximum, FWHM) determines relative crystallite sizes and the presence of defects.[36] The

intensity of a peak, indicating the amount of scattered radiation detected at 20, defines its height.

In materials where interplanar strain is assumed negligible, the Scherrer equation may be used to

determine average crystallite size (D) as follows

D = KX / p cos 0 (2-3)

where K is the Scherrer constant (0.9 for HAP), X is the x-ray wavelength, 0 is the conditioned

FWHM breadth of the diffraction peak and 0 is the diffraction angle for a specified diffraction

peak. Broadening of a diffraction peak suggests a decrease in average crystallite size and/or

increase in lattice defects.[36] Factors such as the multiplicity effect of material planes of the same

family encountering the incident x-ray beam at the same diffraction angle, the structural effect of

the material atomic positions on detected peak discernment and the polarization effect of x-ray

wave propagation in relation to diffractometer detector position must be considered.[35]

Conventional 0-20 XRD studies have been reported in the literature in the evaluation of

untreated and demineralized enamel. Gawda et al. [37 positively correlated the velocity derived

from ultrasonic measurements to crystallite size in dental tissue using XRD. The average

crystallite size in enamel in the c-axis direction was found to be 45.4 5.2 nm and 26.0 4.4 nm

in the a-plane. Crystallinity data were calculated from the (002) and (310) reflections,

respectively. Dentin crystallite size was reported as 23.9 13.0 nm along the c-axis and

6.6 1.2 nm within the a-plane. It should be noted that sample slabs were cut such that they

contained overlapping enamel, dentin and cementum portions of the tooth, thereby affecting the

true calculated crystallinity of each designated tissue component. In an attempt to characterize









LIST OF REFERENCES

[1] U. S. Dept of Health and Human Services, Oral health in america: a report of the
surgeon general, National Institute of Dental and Cranialofacial Research, National
Institutes of Health, Rockville, MD 2000.

[2] U. of Texas Health Sci Center at S. Antonio, in ScienceDaily, 2008.

[3] B. T. Amaechi, S. M. Higham, JDent 2005, 33, 243.

[4] J. D. Bader, D. A. Shugars, JEvidBase Dent Pract 2006, 6, 91.

[5] C. Murdoch-Kinch, M. McLean, JAm Dent Assoc 2003, 134, 87.

[6] J. Bader, D. Shugars, A. Bonito, JPublic Health Dent 2002, 62, 2001.

[7] R. Tang, L. Wang, C. Orme, T. Bonstein, P. Bush, G. Nancollas, Angew Chem Int Ed
2004, 116, 2751.

[8] B. Gillings, M. Buonocore, JDent Res 1961, 40, 105.

[9] A. R. ten Cate, Oral Histology: Development, Structure and Function, The C. V. Mosby
Company, St. Louis 1989.

[10] N. Harris, F. Garcia-Godoy, Primary Preventative Dentistry Appleton & Lange,
Stamford 1999.

[11] M. Mathew, S. Takagi, JRes Natl nst Stand Technol 2001, 106, 1035.

[12] L. Moss-Salentijn, M. Hendericks-Klyvert, Dental and Oral Tissues An Introduction,
Lea & Febiger, Philadelphia 1990.

[13] T. Aoba, AnatRec 1996, 245, 208.

[14] O. Fejerskov, E. Kidd, Dental Caries: The Disease and its Clinical Management,
Blackwell Munksgaard, Oxford 2003.

[15] F. C. M. Driessans, MineralAspects ofDentistry, Vol. 10, Kargar, Basel, Switzerland
1982.

[16] T. Attin, W. Buchalla, M. Gollner, E. Hellwig, Caries Res 2000, 34, 48.

[17] M. Eisenburger, J. Hughes, N. X. West, K. D. Jandt, M. Addy, Caries Res 2000, 34, 289.

[18] M. Finke, J. A. Hughes, D. M. Parker, K. D. Jandt, SurfSci 2001, 491, 456.

[19] J. t. C. Arends, J M, JCryst GI ,, th 1981, 53, 135.












140

120

100 -

80

60

40

20

0


As-Received 10 min EDTA


60 min EDTA


Figure 5- 12. Comparison of as-received and EDTA-treated enamel elastic modulus values. Error
bars represent 95% confidence.


8142-


0202 07 IJ0043hys .-
7000-- 02-02 07EDTA2 0043 h1 _
0--i43 [101 1 1 hll) -l ro,
x' 0 / -

sco-
5000-








200 C.



WO
c --- -- --- --- _---- --- ----___
.- I J _----- -.. ..---... .. ---- ---- ---- -


)0 0 ILU0 0


V,0 1ZO 0 Z 50 0 KnjO U 3A O 40UJ 0
Displacement (Mmi)


Figure 5- 13. Comparison of representative force (load)-displacement curves for as-received and
EDTA-treated enamel. A) As-received. B) 10 min EDTA treatment. C) 60 min EDTA
treatment.





95


4 742


I


__Bt


oo









decomposition correlated with morphological changes along the tooth surface layers should

prove beneficial. The process of enamel remineralization, primarily driven by the use of fluoride-

containing topical treatments, refers to the "reconstruction" of partially-decomposed enamel.

There is much controversy over the underlying mechanisms surrounding this process. Debate

most often ensues among the theories of actual enamel mineral regeneration, the formation of a

mineral-like substance in place of the decomposed enamel, or enamel strengthening through

means other than mineral enhancement. Analysis of the submicron-level changes of the enamel

mineral under both demineralization and remineralization conditions may help to explain

crystallinity effects of these processes within surface enamel. Such an understanding from the

smallest structural unit of enamel to the macroscale may prove beneficial in serving as an aid in

the future development of preventative treatments.

Outline of Dissertation

This research attempts to address the aforementioned concerns posed by the scientific

community regarding the earliest stages of tooth decay, with a focus on activity within the

enamel surface layers. Chapter 2 describes the overall structure of the tooth and material

properties specific to enamel. The dynamic processes of enamel dissolution and regeneration-

namely demineralization and remineralization-are explained. A survey of current

characterization techniques used to assess changes in enamel structural integrity-such as

microradiography, profilometry, microhardness testing and conventional x-ray diffraction

studies-was performed. The tooth decay process is governed by enamel-oral fluid interactions

at the atomic level. As such, current diagnostic techniques lack the resolution necessary to

quantify changes in the surface enamel crystal structure due to early demineralization. The

characterization of chemically-modified enamel on the submicron scale is needed. Chapter 3

outlines the general experimental plan for this research, including enamel surface sample









CHAPTER 2
LITERATURE REVIEW

Tooth Structure

Dental enamel is the hard, protective covering of the crown of the tooth. It is the most

highly mineralized of human tissues, surpassing bone. Apatite, a crystalline calcium phosphate,

comprises 96 wt% of its mineral content. Structurally similar to hydroxyapatite (HAP),

Calo(P04)6(OH)2, enamel crystallites are hexagonal in shape and elongated along their c-axes

(Figure 2-1). Bundles of crystallites form rows of densely-packed rods positioned perpendicular

to the tooth's surface (Figure 2-2). The crystallites are oriented parallel to the vertical axis of the

enamel rod, and fan out to angles of 10-40 along the periphery.[7] The enamel rods measure

4 to 7 tm in diameter, and are layered to give an overall enamel thickness of 1 to 2 mm

(Figure 2-3).[8] The remaining 4 wt% of enamel consists of a protein- and polysaccharide-based

matter intertwined in a fine network among the rods and between crystallites, and water.[9]

Enamel may be considered a composite material, consisting of ordered arrays of "fiber-like"

crystals held together by a permeable organic matrix.

Beneath the enamel layer is the softer, more porous, yellowish dentin. Like enamel, it

consists of a HAP-like mineral phase comprised of smaller crystallites. This tissue is most

similar in structure to bone. Dentin contains only 68 wt% apatite. An organic matrix of

collagenous proteins surrounds the mineral, serving as the structural backbone of this phase.[10]

These microstructural characteristics render dentin more reactive and susceptible to chemical

attack relative to enamel. Dentin, the most abundant dental tissue, determines the size and shape

of the tooth (Figure 2-3). The dentinoenamel junction (DEJ) serves as an interface between the

dentin and enamel. The innervated pulpal cavity is found at the center of the tooth. The root of

the tooth is protected by an extension of enamel, the cementum, within the oral gingiva.









supporting the occurrence of remineralization as measured on the micron scale, there is little

evidence that this phenomenon occurs on the submicron scale.

When discussing remineralization, the influence of fluoride cannot be ignored. Low

concentrations ofF added to calcium- and phosphate-rich remineralization solutions, on the sub-

ppm scale, have been reported to encourage enamel mineral regeneration.[137] Other studies have

shown that the inclusion of 0.1 ppm F in solution during remineralization has prompted calcium

uptake, but no enhancing effects ofF on demineralized enamel were detected.[138] The

remineralization of the inner subsurface caries lesion zone and outer surface layer was reported

at F concentrations within the range of 2 tolO ppm.181l TEM imaging studies have proven that an

increase in enamel crystallite size occurs when subjected to the aforementioned solution types,

and this effect is claimed to be influenced by the presence of F.[41] The "filling" of centrally-

hollowed demineralized crystallites with apatite upon remineralization was also demonstrated,

suggesting mineral enhancement on the submicron level.[711 These results have been corroborated

by XRD studies, which cite an increase in crystallite c-axis length and a-axis width

dimensions.[57] While F does appear to have an enhancing effect on the regeneration of enamel

mineral, is it a necessary component or does it act as a catalyst for the remineralization process?

In this study, selected EDTA-treated enamel slabs were subjected to MI PasteTM and a

synthetic calcifying solution (0.1 ppm F) to illicit enamel surface reconstruction. Resultant

surface enamel microstructural changes were examined by AFM and SEM. Crystallinity studies

were performed using GIXD techniques to analyze the demineralized surface layers of enamel.

Nanoindentation provided a relative comparison of hardness and elastic moduli data relative to

the data calculated from EDTA-demineralized enamel. Raman data, correlate with confocal

imaging observations, and provide relative chemical information within the remineralized









treatment. Sample preparation consisted of dehydrating and coating all samples with Au-Pd

alloy. Secondary and backscattered electron images were recorded with an accelerating voltage

of 5 kV. Images were taken in areas highlighting the effects of demineralization and

remineralization on the enamel rods, the prominent structural components of tooth enamel.

Additional changes in enamel microstructure were imaged with a JEOL JSM-6335F

field-emission SEM (FE-SEM). Its high-density electron beam at a relatively small spot size

provided maximum brightness to discern features as small as 5 nm at 5 kV.

Atomic Force Microscopy (AFM)

For imaging as-received and select treated samples, a Dimension 3100 AFM

(Veeco, Plainview, NY) was used. Since coating the sample is not necessary for AFM analyses,

better contrast and resolution of finely-polished enamel samples could be realized. For this

research, the AFM was operated in the tapping mode. Average RMS roughness values were

extracted in representative areas along select samples. All analyses were performed under

atmospheric conditions using standard Si3Ni4 tips.

Attempted TEM Characterization

TEM analyses of as-received and chemically-treated enamel were attempted for this

research. Due to difficulties encountered in sample preparation for this method, results were not

obtained. The cross-sectional examination of intact enamel from the surface toward the bulk was

desired. A depth profiling of changes in electron density and/or lattice mismatch would have

been correlated with high-resolution x-ray scattering data. Energy dispersive spectrometry (EDS)

were to be attempted with the surface region of the enamel samples to more accurately pinpoint

any phase changes that may have occurred during enamel treatment.

Although the preparation of thin enamel samples for TEM analysis have been prepared

most often through ultramicrotomy, the high deformation induced often leads to cracking of the












1800 -

1500 -

1200 -

900-

600 -




0-
1


Calcifying Soln- h As-Received 30 min EDTA



8








A


0 15 20 25 30 35 40 45 50


02Theta


Figure 6- 20. GIXD spectra comparison of as-received, EDTA- and calcifying solution-treated
enamel at co = 0.40. A) As-received. B) 0.12M EDTA treatment for 30 min. C)
Calcifying soln treatment for 8 h.

80
0 1 --As-Received ---60 min
70


0.15 0.2 0.25 0.3 0.4
Grazing Incidence Angle, o (0)


0.8 1.0


Figure 6- 21. Representative depth profile of the average enamel crystallite size along the c-axis
as a function of co for as-received and 60 min EDTA-treated enamel, ending with a 5
min MI PasteTM and a 8 h calcifying solution treatment.










70
S60
S0 R= 0.9541
3 50 -
4- 40-
O 30 -
0 20
o /


20 23 25 28 30 33 35 38 40

Crystallite Size (nm)


Figure 5- 9. Correlation of EDTA treatment time with crystallite size at co = 0.15.


1.0E+06


1.0E+05

1.0E+04

1.0E+03


1.0E+02


1.OE+01 I 1 '"
0.2 0.5 0.7 1.0 1.2 1.5 1.7 2.0
02Theta


Figure 5- 10. Representative XRR curve comparison of as-received and EDTA-treated (30 min)
enamel.









Raman spectrum extracted from the X-Y scan showed rather sharp vi- V4 phosphate bands in the

lower spectral region, in addition to bands indicative of organic material at the higher

wavenumber value ranges (Figure 6-16). The C-H stretching band present in the as-received

Raman spectrum, which may be shifted in the EDTA-treated Raman spectra, was replaced by a

stronger organic material presence. This is indicated by the relatively, well-defined peaks toward

the higher end of the spectrum.

Buffered lactic acid-treated enamel substrate

AFM images of a 5-min MI PasteTM treatment on buffered lactic-acid-demineralized

enamel in Figure 6-17 show granular patches of matter dispersed randomly along the sample

surface. Enamel rod outlines were faintly detected in some areas. An RMS roughness value of

99.4 nm was calculated over the entire image, perhaps primarily due to the two large humps

observed on the surface. Figure 6-18 shows a decrease in crystallite size in the [002] direction

closest to the surface, with a marked increase in crystallinity at depths greater than 120 nm.

Calcifying Solution Treatments

EDTA-treated enamel substrate

Figure 6-19 shows AFM images of demineralized enamel (0.12M EDTA for 30 min) after

an 8 h treatment with the calcifying solution. The demineralized enamel microstructure was still

fairly visible, with raised patches of matter randomly distributed along the surface. An averaged

RMS roughness value of 408 nm was calculated over the entire image.

A representative GIXD spectra comparison (co = 0.40) of EDTA-demineralized enamel

immersed for 8 h in a calcifying solution showed such an overall decrease in peak intensity such

that the peaks could barely be discerned (Figure 6-20). There was also a decrease in peak

intensity within the halo region. A representative depth profile graph of crystallite size changes

from the surface toward bulk enamel in Figure 6-21 showed a negligible change in crystallinity

109









layer can be exacted from the 6 and 6, values calculated from Equations (3-2) and (3-3),

respectively. [100]

Grazing incidence x-ray diffraction (GIXD)

During GIXD, the x-ray beam penetrates the sample a few thousand angstroms, irradiating

a large area along its surface. In contrast to conventional 0-20 and similar to XRR studies, the

beam is directed toward the sample at a small incidence angle (co) around Oc. A portion of the

beam is diffracted (not reflected) by the sample and is used to generate GIXD patterns of the

surface layers of the material.

With a GIXD measurement, co remains constant as the x-ray detector moves in the plane of

26. Varying co for each measurement creates a depth profile of crystallinity (peak) information

within the surface region (Figure 3-5). A relatively large co value (above -7.00) above the critical

angle will cause x-ray penetration into the bulk, in which case the bulk diffraction peaks may

obstruct those of smaller intensity from the surface. The GIXD geometry does not allow the

presence of strong diffraction lines from the sample substrate (i.e., the bulk) from interfering

with the spectra of the sample itself. Figure 3-6 displays its strength in resolving surface layer

crystallinity in titania deposited onto a glass substrate. The amorphous nature of the glass

detected by conventional XRD methods produced a halo that overshadowed the crystallinity data

exploited by GIXD in the same region. GIXD is capable of detecting crystallinity changes at

depths of 10-100 nm from the sample surface.

Experimental set-up

In this research, copper Ka radiation (X = 1.542 A) was used. All measurements were made

as continuous scans in the powder mode. The machine operated at 45 kV and 40 mA under the

line focus of the x-ray beam. The Cu K, attenuation length for HAP, which describes the depths

from which diffraction data are extracted, is shown in Figure 3-7. For 0-20 measurements, a 1/20









of the sample. A smaller height measurement was an indication of enamel loss. Baseline

measurements were taken after minimal polishing of the samples with a combination of 180 and

240 grit papers. The best resolution available from the penetrometer was 0.1 mm. The samples

were also visually inspected prior to each measurement to check for exposure of the underlying

dentin tissue.

Prior to characterization, the enamel slabs were additionally subjected to an alumina slurry

polishing regimen as outlined in Chapter 3. It was determined that enamel loss due to slurry

polishing was too small to be detected by the penetrometer. As such, those measurements were

omitted from this study.

Results

Relative sample thicknesses of epoxied samples are listed in Table A-1. An average

enamel loss of 0.4 + 0.075 mm was calculated. Upon visual inspection, all samples appeared to

expose enamel only at the surface.

Discussion

As the average mineral loss detected was much greater than 100 gm (0.1 mm), this

polishing regimen seems suitable for the removal of fluoridated enamel from the sample surface.

In addition to F content, overall enamel thickness varies from tooth-to-tooth. The possible

exposure of dentin is a concern, and must be monitored carefully. In addition to visual surface

inspections, cross-sectional examinations of the remaining enamel thickness of each sample

should be considered in this research.









ACKNOWLEDGMENTS

First, I would like to thank God for His blessings and never-ending presence in my life. I

would like to thank my family and friends for the endless support, encouragement and love

provided throughout the years, and especially throughout the dissertation process.

I would also like to thank my dissertation chair, Dr. Valentin Craciun, as well as my

committee members-Dr. Amelia Dempere, Dr. Kenneth Anusavice, Dr. Susan Sinnott and Dr.

Yasumasa Takano. They have all enriched my graduate school experience in various ways-

from engaged discussions of the research, financial assistance for business travel and

conferences, general encouragement during rough patches to agreeing to serve as members of my

supervisory committee. Special thanks go to Mr. Ben Lee and Mr. Nai-Zheng Zhang for

assistance with sample and experimental solution preparation. The two of you always made me

feel welcome during the time I spent in the Dental labs. I would also like to thank Drs. Luis and

Barbara Muga for kindly sterilizing teeth for my experiments and for the great conversations.

Data obtained for my research would not have been possible without the help of the MAIC

staff-thank you all. I would also like to express appreciation for my labmates, Chuchai and

Junghun, for their encouragement and help with sample characterization.

Lastly, I would like to thank the UF Materials Science and Engineering department for the

opportunity to pursue my research and providing opportunities for financial assistance.









similar to a caries lesion white spot. Microstructurally, the enamel rod pattern is exposed, as seen

by electron microscopy (Figure 2-9). Etched lesions typically extend to depths of 10 inm. The

increased surface area of etched dental tissue proves ideal for tooth bonding treatments, which

are often initiated with phosphoric or ethylenediamine tetraacetic acid (EDTA).[21] Treatment

times may vary from a few seconds to several minutes, and is highly dependent upon the etchant

used.

While the chemical processes surrounding enamel dissolution are well-defined, there is a

lack of information concerning the structural considerations involved-particularly along the

enamel surface. This may be due to the limited means available to access structural and chemical

changes in tooth enamel at its most basic structural level-along the nanoscale. Current methods

for the detection of tooth decay must be considered, as these are the primary ones used to

monitor changes in enamel as a result of dissolution.

Microradiography (MR)

The use of x-ray absorption to examine the mineral structure of dental enamel was

introduced by J. Thewlis[22] in 1940. This concept has evolved into the MR methods used today,

long regarded as the standard means to assess dissolution of hard dental tissue. The most popular

form of MR for decay detection is transverse microradiography (TMR). A CuK, x-ray beam is

directed toward a thinly-sliced sample (<100 [tm for enamel) and a nearby calibration wedge.

This set-up is enclosed in a light-proof casing, with radiographic film placed beneath the sample

and wedge.[23] Monochromatic x-rays interact with the sample, wedge and film to produce a

microradiographic image. Microdensitometry tracing of the image allows calculations of the gray

levels present in comparison to a standard. Lighter areas on the micrograph indicate intact

enamel; darker areas correspond to those of deficient mineral content. Data indicating enamel

mineral loss (AZ) and decay lesion depth (Ld) can be extracted. The mineral volume %, V(x), at










6

5

4

3

2

1-

0
As-Received Lactic Acid 370C Lactic Acid 200C


Figure 5- 24. Comparison of as-received and buffered lactic acid-treated enamel nanohardness
values. Error bars represent 95% confidence.






120

100

80

60

40

20 -

0
As-Received Lactic Acid 370C Lactic Acid 200C


Figure 5- 25. Comparison of as-received and EDTA-treated enamel elastic modulus values. Error
bars represent 95% confidence.









LIST OF TABLES


Table page

2- 1. Minor inorganic constituents in sound human tooth enamel ...........................................37

2- 2. Sampling of sound tooth enamel hardness and modulus values. ........................................41

3-1. Composition of buffered lactic acid demineralization solution ................ ......... ..........58

3-2. Composition of calcifying remineralization solution.................................................58

3-3. Raman frequencies and P043- vibrational modes associated with tooth enamel....................61

5-1. Comparison of as-received and demineralized enamel hardness and modulus values. .........94

6-1. Comparison of hardness and elastic modulus values of as-received, EDTA-
demineralized and enamel remineralized with MI PasteTM and a calciyfing solution
(0 .1 p p m F ).......................................................................... 12 1









divergence slit was used. A step size of 0.020 and time per step of 7.0 s were used to produce a

high signal to noise ratio. For GIXD, the same scan parameters were used, except for the use of a

1/80 divergence slit. In addition, a succession of co values between 0.150 and 0.40 were used to

monitor changes near the surface of the samples. Grazing incidence beam values of 0.8 and 1.0

were also run and extrapolated to 0-20 to trace crystallinity changes into bulk enamel. Portions of

the resultant spectra were modeled with ProFit curve fitting software.[101] For XRR

measurements, a 1/320 divergence and 0.1 mm receiving slits were used. A step size of 0.0050

and time per step of 6.0 s within the scan range of interest were used.

Nanoindentation

Nanomechanical testing on tooth enamel can be used to quantify changes in surface

structure. A diamond indenter is typically used for probing. The probe contacts the sample

surface with a predetermined force at a particular rate. A force-displacement curve is generated,

which can be used to determine the mechanical properties of the sample.[69] Using a power law

relationship, the applied load (P) is determined as

P = A(ho-hf)m (3-7)

where A is the surface area, ho as the initial penetration depth, hf as the final depth, and m = 2

(conical). The stiffness (S) of the material is defined as the derivative of the load with respect to

the penetration depth at the maximum load, Pmax

S = dP / dh at Pmax (3-8)

The contact depth (he) of the probe at a particular instant is defined as

hc= hmax 0.75(Pmax / S) (3-9)

Finally, the hardness (H) and reduced modulus (Er) values were found by

H = Pmax / Ahe (3-10)

Er = [S 71t] [2 ~(Ahe)] (3-11)




























Figure 6- 3. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI PasteTM for
5 min, highlighting a single enamel rod.


Figure 6- 4. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI PasteTM for
5 min, highlighting central region of a single enamel rod.

































Figure 5- 3. FE-SEM image of as-received enamel treated with 0.12M EDTA for 10 min.


Figure 5- 4. FE-SEM image of as-received enamel treated with 0.12M EDTA for 60 min.


MAIC :I I IkV 14,;IiIrII









acidic exposure until the demineralization effects spread to the periphery of the crystal, leaving

mere fragments of mineral behind. Structural integrity is completely compromised at this point.

Untreated sections of tooth enamel examined by TEM exhibited a lattice structure similar

to stoichiometric HAP in applicable studies. Bres and Hutchison[42] suggested the influence of

enamel matrix proteins on crystallite growth. They found that growth control by the matrix

proteins only occurs on the {120} planes. An examination minan of the matrix showed the presence of

poorly-crystalline, HAP-like phases, which were shown to form grain boundaries with the

enamel crystallites.

While the aforementioned techniques provide insight into mineral changes that occur as a

result of enamel dissolution, they are inadequate in examining such changes representative of

that at the tooth-oral fluid interface. Higher resolution surface analysis techniques would prove

more helpful, and may assist in treatment of tooth decay. As with bone, albeit through different

mechanisms, partially-decomposed enamel may be regenerated. This process has been termed

"remineralization," and may occur in carious, and in some cases eroded and etched enamel.

Enamel Remineralization

After periods of demineralization, in vivo, the tooth has a propensity to naturally repair

itself. Partially-dissolved enamel has the ability to be regenerated to a near-healthy state via the

transformation of enamel crystals. Partial enamel dissolution that leaves behind a suitable

mineral substrate act as nucleation sites for growth. When the bulk saliva pH rises above the

critical pH for enamel dissolution, there is a shift in equilibrium whereby the saliva becomes

supersaturated with Ca2+ and P043-.[14] Demineralization wanes and remineralization dominates,

driving the calcium and phosphate ions to diffuse into the decomposed enamel to form new or

modified apatite. The dynamic, interchangeable, and sometimes concurrent processes of

demineralization and remineralization in vivo conduct a balancing act in reacting to disruptions









Similarly, frequent or prolonged exposure to acidic agents that work to induce enamel

erosion is a recurring problem. Data collected from San Antonio, TX, as part of an on-going

nationwide study, suggested a 30% prevalence rate among preteens.[2] With detected erosion

rates of nearly 50% in Western European countries among the teenage and younger populations,

erosion is gaining attention in the area of dental pathology.[3]

Clinical Concerns

Although the advent of fluoridated drinking water and topical fluoride treatments has

tremendously improved the state of oral health worldwide, the statistics suggest that more work

is needed. The depressed state of oral health in America has prompted a shift in focus from

reactive to preventative dental care.141 The earlier that enamel dissolution can be detected and

treated by cost-effective, minimally-invasive means, the greater the chance for preservation of

the structural integrity and functionality of the tooth. This proposition is especially relevant for

residents in low socioeconomic environments, who often have limited access to the resources

necessary to combat the advanced stages of tooth decay. Progress in early caries treatments,

materials selection for tooth restoration and an increased understanding of the overall caries

development process has made preventive dental care a possibility.[5] Interestingly, reports have

indicated that clinicians are hesitant to adopt this strategy in practice. There is weak evidence

supporting the ability of current diagnostic techniques to accurately detect the earliest stages of

tooth decay. A review performed for the 2001 NIH Consensus Development Conference on

caries management suggested that several diagnostic tests-including microradiographic, laser

fluorescence and electrical conductance methods-yielded results that were largely inconclusive,

with great variance due to operator error and data relevance. As such, a baseline protocol to

clinically assess early stages of tooth decay has yet to be developed.[6] An increased

understanding of the crystallinity, chemical and mechanical changes resulting from early enamel









treatment time of 30 min was selected as it has been cited as being the amount of time needed for

acid clearance in the mouth after a meal.1491 Demineralization effects were intended to represent

those seen along the enamel surface as a result of early caries lesion development.

Remineralization

Calcifying solution

Saliva is known to aid in the remineralization of enamel through its buffering effect on

acids found in the oral environment and when supersaturated with calcium and phosphate with

respect to demineralized enamel. It also contains trace amounts of fluoride (< 0.1 ppm), which is

widely accepted as a means to shift the decay process to favor remineralization and to facilitate

the diffusion of calcium and phosphate ions from the saliva back into the enamel crystal

structure. 52, 85, 86] This action may also encourage the production of FAP.187' 88] The use of

artificial saliva or calcifying solutions in vitro has introduced the saliva effect into the evaluation

of eroded and caries-induced demineralization. These solutions often contain an organic buffer,

such as N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), and supersaturated

concentrations of calcium and phosphate.

In this study, a calcifying remineralization solution[511 (pH = 6.97; [F] = 0.1 ppm) was used

to simulate artificial saliva effects in vitro (Table 3-2). As-received enamel slabs were immersed

in at least 25 ml of solution for up to 8 h, without agitation. The long treatment time was chosen

to simulate the remineralization effects of the saliva overnight.

MI PasteTM topical cream

Topical fluoride treatments-namely dentifrices, mouthrinses and gels-have proven

effective in protecting the tooth against enamel dissolution. Containing relatively high

concentrations of fluoride (up to tens of thousands of ppm), some treatments have the ability to

form a calcium fluoride-like substance in plaque and on surface enamel to provide a reservoir of


































2008 Carmen Veronica Gaines









oral care over the lifetime of the teeth. Lateral and depth variability in enamel thickness,

hardness and crystallinity for each enamel sample must be considered during data analysis.

In addition to inherent structural variability, experimental environment plays an important

role. Whole teeth thrive in a warm (i.e., human body temperature of 37 C), moist environment

in vivo. Deviations from the tooth's natural state were incorporated into this research. All

experimentation was performed in vitro and at room temperature, unless otherwise specified.

Due to high vacuum needs of the SEM, the enamel samples were dehydrated prior to coating.

This made the enamel samples more susceptible to cracking. Enamel polishing was necessary to

provide a relatively large, flat surface area with minimal roughness, important for instrument

error reduction in high-resolution XRD and nanoindentation studies. Polishing also allowed for

better standardization of the enamel studied in that the natural curvature of the incisor teeth was

flattened and outer layers of fluoridated enamel was removed. Unfortunately, polishing induces

physical trauma to the tooth, introducing possible defects. Finally, tooth storage methods after

polishing and prior to characterization may affect enamel material properties. Reports indicated

that the storage of tooth enamel sections in deionized water causes a decrease of its

nanomechanical properties by 50% after 1 week of exposure, as determined by nanoindentation

studies.[120] Diffusion processes extracted the structurally-important Ca2+ and P043- ions from the

enamel into the water, causing demineralization. As such, the enamel samples were treated

and/or characterized shortly after polishing to decrease idle storage time. When stored, the teeth

were kept hydrated, although not fully immersed, in distilled water.









brittle enamel matter. To help eliminate this problem, focused ion beam (FIB) milling has been a

proposed method for sample thinning. FIB has been attempted on synthetic apatite materials,11071

and recently on actual tooth enamell108' 109] and dentin1110 111] tissue. Although HAP is cited as

being resistant to ion beam damage, less stable apatitic stages may amorphize.11121 This condition

may alter the chemistry of remineralized enamel surfaces. As the FIB process is

time-consuming and takes great precision and accuracy, we had difficulty developing a sample

preparation method compatible with the Strata DB235 (FEI, Hillsboro, OR) system available for

use. Due to lack of FIB sample availability, TEM analysis was not attempted.

Statistical Considerations

Due to the limited number of suitable teeth available for analysis, significant statistical

analyses could not be achieved. For crystallite size calculations, at least three samples from each

treatment stage were analyzed. Results were plotted in a line graph showing an averaged

crystallite size with error bars indicating the standard deviations of the sizes calculated. While

the results may not be statistically relevant, trends in enamel crystallite behavior could be

observed. Differences in between as-received and treated enamel groups analyzed by

nanoindentation and XRD calculation of average crystallite size were determined with 95%

confidence.









[62] F. C. M. Driessans, R. H. M. Verbeeck, Eds., Biominerals, CRC Press, Boca Raton 1990.

[63] R. Z. LeGeros, Calcium phosphates in oral biology & medicine, S Kargar AG, Basel,
Switzerland 1991.

[64] R. H. Selwitz, A. I. Ismail, N. B. Pitts, Lancet 2007, 369, 51.

[65] M. Takahashi, J. Zheng, K. Mori, I. Mataga, K. Kobayashi, Materials Science andEng C
2006, 26, 703.

[66] H. Komatsu, H. Yamamoto, M. Namachi, K. Yasuda, Y. Matsuda, Y. Murata, T.
Kijimura, H. Sano, T. Sakai, T. Kamiya, NuclInstr andMeth in Phys Res B 2007, 260,
201.

[67] M. E. Barbour, J. S. Rees, JDent 2004, 32, 591.

[68] D. M. Ebenstein, L. A. Pruitt, Nanotoday 2006, 1, 26.

[69] B. Bhushan, Ed. Nanotribology and Nanomechanics An Introduction, Springer-Verlag,
Berlin, Germany 2005.

[70] C. Nave, HyperPhysics, http://hyperphysics.phy-astr.gsu.edu/hbase/quantum/bragg.html,
March 2008

[71] H. Tohda, T. Yanagisawa, N. Tanaka, S. Takuma, JElectron Microsc 1990, 39, 238.

[72] J. M. White, JDent Res 1994, 73, 1560.

[73] Y. Iijima, O. Takagi, J. Ruben, J. Arends, Caries Res 1999, 33, 206.

[74] J. M. ten Cate, Acta Odontol Scand 1999, 57, 325.

[75] N. W. Johnson, Arch OralBiol 1966, 11, 1421.

[76] J. H. Meurman, J. M. ten Cate, Eur J Oral Sci 1996, 104, 199.

[77] P. Bogra, S. Kaswan, JIndian Soc Pedo Prey Dent 2003, 21, 79.

[78] K. Kawasaki, J. Ruben, I. Stokroos, O. Takagi, J. Arends, Caries Res 1999, 33, 275.

[79] Z. C. Cehreli, N. Altay, Angle Orthodontist 2000, 70, 22.

[80] J. W. Simmelink, V. K. Nygaard, D. B. Scott, Arch OralBiol 1974, 19, 183.


[81] H. Yamazaki, A. Litman, H. C. Margolis, Arch Oral Biol 2007, 52, 110.









CHAPTER 4
ENAMEL CHARACTERIZATION

Introduction

An adequate assessment of changes in surface-treated enamel requires comparisons to

sound, untreated enamel. Characterization of such a sample provides a baseline for relative

comparisons of structural, mechanical and chemical changes in the enamel upon

demineralization and remineralization. The microstructure of enamel, with enamel crystallite

dimensions as small as tens of nanometers, requires careful consideration and precision in

describing its material properties.

The hierarchal microstructure of enamel results in anisotropic mechanical and structural

properties due primarily to its preferred crystallite/enamel rod orientation. Enamel crystallites-

bundled to form enamel rods-are orthogonally-positioned relative to the outer tooth surface. A

pliable, organic matrix cushions the rods and absorbs compressive forces induced by chewing

and biting. In essence, enamel is analogous to a more generalized description of a fiber-

reinforced composite.1114, 115] The literature cites evaluations of mechanical properties parallel

and orthogonal to the occlusal (biting) surface of human molars, schematically represented in

Figure 4-1. Cuy et al. [116] found slight variations in hardness and elastic modulus along the tooth

surface, with the highest values measured along the occlusal surface. Interestingly, variation

from the enamel surface to DEJ varied greatly. Mechanical property dependence on orientation

provides an example of the influence of the crystal structure of tooth enamel on tooth form and

function, a topic explored throughout this research.

In addition to local variance in mechanical property values, changes in crystallite size and

overall enamel thickness with position were also reported. Low[98] reported that enamel had a

graded nature in terms of crystallite size, with an average decrease in size from the tooth surface

































Figure 2- 15. TEM image of cross-sections of demineralized enamel crystallite c- axes.[71]


Demineralization


mineralization


Oral bacteria
Fermentable carbs
Susceptible enamel surface
Time


Readily-available Ca POP-, F sources
Adequate salivary flow
Low sugar/addic diet


Figure 2- 16. Schematic illustrating the balance of demineralization and remineralization
processes with regard to tooth decay.[64]
































21 21.5
02Theta


22 22.5


Figure 6- 9. (002) GIXD spectra comparison for as-received, EDTA-treated (30 min) and
MI PasteTM-treated (5 and 10 min) enamel at co = 0.40.


80

70 +--As-Received ---60 min
--MI Paste 0 h -MI Paste 24 h
S60

S50

40

S30

20

10


0.15 0.2 0.25 0.3 0.4 0.8 1.0
Grazing Incidence Angle, o (o)


Figure 6- 10. Representative depth profile of the average enamel crystallite size along the c-axis
as a function of co for as-received enamel treated with EDTA for 60 min, ending with
a 5 min MI PasteTM treatment. Treatment stability highlighted upon immersion in
distilled water for 24 h.


1100


500 -t1



300 -
20.5









Subsequent remineralization evaluations performed by this research will focus on apatite

"regeneration" effects on EDTA-treated enamel, due to the greater degree of material property

changes observed upon demineralization. Examples of the remineralization of buffered lactic

acid-treated enamel substrates will be included for comparison purposes.









after a 60 min EDTA treatment (Figure 5-5). The emergence of the (211) reflection was seen as a

shoulder on the (112) reflection, a peak typically resolved in bulk enamel XRD spectra. The

latter occurrence was not seen during the relatively short, 10 min treatment. In a single

experiment, crystallite size in the [002] direction increased from 25.56 nm to 28.71 nm as the

EDTA treatment time increased from 10 min to 60 min. These results suggested a change in the

near-surface enamel crystal structure under the aforementioned experimental conditions, an

observation with a direct influence upon subsequent experimental design. As mechanical

property and topographical evaluations of these samples were not performed, the potential

resistance of apatite dissolution after the 60 min treatment time could not be verified. In an

attempt to evaluate enamel dissolution effects only, future experimental conditions were adjusted

such that a greater volume of treatment solution was used.

A subsequent dissolution experiment on a representative EDTA-treated sample showed

GIXD spectra depth profile comparisons (co = 0.40) of as-received enamel and that treated for

30 and 60 min. A progressive slight sharpening of the enamel peaks with treatment time was

observed (Figure 5-6). The emergence of the (211) reflection after the 60 min treatment was not

detected. More obvious progressive improvements in crystallinity were seen with the (112)

reflections. Figures 5-7 shows an averaged depth profile of changes in enamel crystallite size

from the surface toward bulk enamel. An increase in crystallite size with treatment time was

observed closest to the enamel surface (Figure 5-8). The greatest changes in crystallite size were

calculated with the 60 min treatment closest to the enamel surface at co = 0.150. There was a

positive correlation (R2 = 0.9541) between enamel crystallite size and EDTA treatment duration

per the conditions of this experiment (Figure 5-9). The crystallite sizes appeared to converge to a

smaller value range closer to bulk enamel, suggesting that demineralization had negligible









Confocal Raman studies of as-received enamel showed an X-Y surface scan consisting of

an array of enamel rod outlines. They appeared to correspond to the terminal ends of the

vertically-aligned enamel rods seen in cross-section in Figure 4-10. The images were color-

coded in orange by averaging the highest-intensity vi(PO) band across the images. The black

areas within the mineral regions of the images represent the organic matrix. The averaged Raman

spectrum indicated the presence of rather sharp phosphate band vibrational peaks at 432/cm,

586/cm, 959/cm and 1070/cm (Figure 4-11). A C-H stretching band, which indicates the

presence of organic material, was seen.

Discussion

Tooth Enamel Defined

FE-SEM imaging showed the terminal ends of relatively intact enamel crystallites, with

surface flattening via polishing working to expose preferential orientation along their c-axis

directions. AFM imaging did not resolve crystallite features due, in part, to the scan size

selected, a decision that will become more obvious in later experimental studies. Crystal

orientation is further corroborated by the presence of only a few reflections of the total number

possible in the 0-20 spectrum of the enamel. This indicates texturing within the material. The

(002) reflection has the highest intensity, indicating that the greatest number of x-rays are

reflected from the (002) planes. This provides further evidence of the crystallite c-axis direction

as the preferred orientation. A lack in overall enamel spectrum intensity as compared with pure

HAP may indicate the relatively poor crystalline quality of enamel and x-ray scattering

interference from the organic material present in enamel. The organic material is indicated by the

presence of a broad amorphous peak, or halo, in the 0-20 spectrum. The benefit of the GIXD

technique is illustrated in Figure 4-5, where the presence of the (200) reflection within surface

enamel is resolved. This peak cannot be discerned with conventional XRD techniques. A

66









components is common.[94-97] As such, the samples used in this research may be analogous to the

structure of a polycrystalline thin film/substrate combination in that the treated surface layer is

representative of a thin film atop an untreated, bulk enamel substrate (Figure 3-3). To accurately

evaluate treated enamel, characterization methods sensitive to surface interactions are desirable.

Various high-resolution characterization techniques were used to describe changes in

surface morphology, structure, mechanical properties and chemistry after the demineralization

and remineralization of tooth enamel. Grazing incidence x-ray diffraction (GIXD) and x-ray

reflectivity (XRR), techniques widely used in thin film analyses, were used to assess structural

changes in the surface layers of enamel upon treatment. High-resolution x-ray scattering results

were corroborated with nanoindentation studies and confocal Raman microscopy performed on

select samples to assess changes in mechanical properties and surface layer chemistry,

respectively. Surface imaging was done using scanning electron microscopy (SEM) and atomic

force microscopy (AFM), performed to examine the topographical features of as-received and

treated enamel samples.

High Resolution X-ray Scattering

Advanced x-ray scattering methods were used to assess crystallinity changes in tooth

enamel as a result of demineralization and remineralization treatments. X-ray diffraction studies

were performed using an X'Pert Pro Materials Research Diffractometer (PANalytical,

Westborough, MA). With regard to this research, its high resolution capabilities were most

suitable for XRR and GIXD phase analyses. Use of the X'Pert provides a nondestructive

evaluation of the structural, and to a degree, chemical characteristics of the material of interest.

The X'Pert Pro uses a four-axis goniometer system that allows for precise sample, beam

and detector placement for material analysis. The 20 parameter, which defines the angle between

the incident and diffracted beams, controls the positioning of the detector. The omega (co)





























Figure 5- 18. AFM images of as-received enamel treated with buffered lactic acid for 30 min at
human body temperature. Scale =10 gm2


Figure 5- 19. SEM images of as-received enamel treated with buffered lactic acid for 30 min at
room temperature. Scale bar = 20 im. A) Secondary electron image. B) Backscattered
electron image.









[41] J. W. Simmelink, S. C. Abrigo, Adv Dent Res 1989, 3, 241.

[42] E. F. Bres, J. L. Hutchison, JBiomedMater Res 2002, 63, 433.

[43] S. Takagi, H. Liao, L. C. Chow, Caries Res 2000, 34, 281.

[44] D. T. Zero, "Dentrifices, mouthwashes, and remineralization/caries arrestment
strategies", presented at Biotechnology and Biomaterials to Reduce the Caries Epidemic,
Seattle, 2005.

[45] B. Ogaard, Caries Res 2001, 35, 40.

[46] H. Duschner, H. Gotz, B. Ogaard, Eur J Oral Sci 1997, 105, 466.

[47] H. Tsuda, W. L. Jongebloed, I. Stokroos, J. Arends, Caries Res 1993, 27, 445.

[48] J. M. ten Cate, JDent Res 1990, 69, 614.

[49] C. Robinson, J. Kirkham, A. C. Baverstock, R. C. Shore, Caries Res 1992, 26, 14.

[50] R. M. 0. Argenta, C. P. M. Tabchoury, J. A. Cury, Pesqui OdontolBras 2003, 17, 241.

[51] J. M. ten Cate, P. P. E. Duijsters, Caries Res 1982, 16, 201.

[52] B. T. Amaechi, S. M. Higham, Arch Oral Biol 2001, 46, 697.

[53] L. M. Silverstone, J. S. Wefel, J Cryst Gi ,, tih 1981, 53, 148.

[54] K. Collys, R. Cleymaet, D. Coomans, Y. Michotte, D. Slop, Caries Res 1993, 27, 15.

[55] F. Lippert, D. M. Parker, K. D. Jandt, JColloidlnterface Sci 2004, 280, 442.

[56] T. Aoba, Y. Moriwaki, Y. Doi, M. Okazaki, J. Takahashi, T. Yagi, JOralPatholMed
1981, 10, 32.

[57] A. Vieria, R. Hancock, H. Limeback, M. Schwartz, M. Grynpas, JDent Res 2003, 82,
909.

[58] H. R. Muhlemann, H. Lenz, K. Rossinsky, Helv OdontActa 1964, 8, 108.

[59] K. Kawasaki, Y. Tanaka, O. Takagi, Arch Oral Biol 2000, 45, 797.

[60] R. Z. Wang, S. Weiner, JBiomech 1997, 31, 135.

[61] F. H. Jones, Surface Science Reports 2001, 42, 75.









errors in deeper layers showing improved crystallinity in the crystallite [002] direction. Due to a

lack of statistically-relevant crystallinity and nanoindentation data, the true effects of MI PasteTM

remineralization cannot be verified.

Calcifying Solution Treatments

Upon remineralization with a calcifying solution (0.1 ppm F), a slight surface layer seemed

to form on both EDTA- and buffered lactic acid-demineralized enamel samples. This assumption

was verified for the EDTA-treated substrate through a GIXD spectra comparison near the surface

of remineralized enamel highlighting the obvious decrease in peak, and amorphous halo,

intensities. For EDTA-demineralized enamel, a decrease in crystallite size below that of as-

received enamel was detected by GIXD. These results suggest that an amorphous surface layer

may have obstructed pertinent peak detection by GIXD. For buffered lactic acid-demineralized

enamel, an increase in crystallite size along the enamel surface was seen. This may be due to

compound effects influenced by F on the relatively well-mineralized substrate produced by the

buffered lactic acid (0.5 ppm F) demineralization, although it is interesting to note that there

appeared to be little change in crystallite size based on SEM observation between the MI

PasteTM- and calcifying solution-treated enamel. Of course, in such an instance, the underlying

demineralization effects on crystallite size cannot be ignored.

Enamel remineralization appears to be a complicated process that depends on a variety of

factors, namely enamel F concentration, the state of demineralized enamel and experimental

environment/conditions. In evaluating remineralization effects through changes in XRD

crystallinity, a crystallite size comparison of as-received and remineralized enamel may be most

appropriate in drawing conclusions about the integrity of tooth enamel upon remineralization, as

this study cannot confirm the validity of the structural effects seen as a result of

demineralization. A more focused study utilizing a greater sample size, utilizing instrumentation

114













95% of intact enamel
mineral contents














5% of intact enamel
mineral contents


40 50

Deptl (pm)


Figure 2- 10. Schematic of an average enamel mineral content MR profile showing the mineral
loss lesion (AZ) and the lesion depth (Ld).[66]



XR taion 2 YV RoMln M04


-i-.lt C",
.-- -004_, .."00









'-- -,' '. .. 0- --40
0 A
















Figure 2- 11. Profilometric trace of eroded enamel lesion by a demineralizing agent for 30 min.
Scale is in gm[67]
.,-, ._ 04 .


Ld(pn)






























Figure 6- 22. AFM images of buffered lactic acid-treated enamel (room temperature) subjected
to a calcifying solution for 8 h. Scale = 20 Crm2


Figure 6- 23. FE-SEM image of 30 min buffered lactic acid-treated enamel (room temperature)
subjected to a calcifying solution for 8 h, highlighting the central region of a single
enamel rod.









CHAPTER 7
CONCLUSIONS

Important Findings from the Research

The purpose of this study was to determine structural changes within the surface layers of

tooth enamel upon demineralization and remineralization on the submicron scale. GIXD

provided a depth profiling of crystallinity changes from the surface to several hundreds of

nanometers into the enamel, a resolution only achieved through TEM methods. The

nanoindentation process was refined for as-received enamel to achieve results comparable to

those cited in the literature. The nanomechanical behavior of treated enamel was compared to

these baseline values to determine relative mechanical effects. These results were corroborated

with well-established morphological imaging and Raman scattering techniques in hopes of

providing greater insight into the mechanisms surrounding early stages of enamel dissolution and

regeneration. Similarly, GIXD was used to examine the nature of the material produced along

surface enamel as a result of remineralization. This study also aimed to determine whether

epitaxial regrowth of apatite occurred on partially-demineralized enamel, as proposed by a

number of researchers based on microradiography, microhardness and conventional 0-20 XRD

studies.

Enamel slabs, cut from intact human incisor teeth, were highly-polished to a relative

mirror-like finish for standardization purposes throughout this study. Through GIXD

experiments, surface enamel was found to have a slightly different crystal structure along its

surface. The emergence of the (200) peak, which is unable to be discerned through conventional

XRD methods, was evident. Upon the treatment of surface enamel with 0.12M EDTA, SEM

imaging showed decomposed regions of distinct enamel rod outlines randomly dispersed along

an otherwise smooth enamel surface after 10 min which progressed into extensive dissolution









with remineralization in the [002] direction beyond that originally detected for as-received

enamel. The effects of the MI PasteTM treatment appeared to be negated. This behavior ensued to

depths of approximately 228 nm.

Results of nanoindentation testing for the calcifying solution remineralization of EDTA-

treated enamel, as compared to previously-reported as-received and demineralized enamel

values, are displayed in Table 6-1. Average nanohardness was significantly higher than that

calculated for the 60-min, EDTA-treated enamel. The average elastic modulus for the

remineralized enamel was an order of magnitude higher than that for the EDTA-treated enamel,

although statistically lower than that calculated after MI PasteTM remineralization. Representative

load vs. displacement curves are shown in Figure 6-13.

Buffered lactic acid-treated enamel substrate

AFM images of an 8-h calcifying solution treatment on buffered lactic acid-demineralized

enamel in Figure 6-22 showed a sample surface of variable height, with irregularly-shaped

globules of lightly-textured matter randomly dispersed. Enamel surface height was less uniform

than in samples evaluated elsewhere for this research. An RMS roughness value of 20 nm was

calculated over the entire image, a value closer to that calculated for buffered lactic acid-treated

enamel at human body temperature. Figure 6-23 shows random aggregation of enamel

crystallites similar to that seen in EDTA-demineralized and MI PasteTM-remineralized enamel. A

representative increase in crystallite size in the [002] direction closest to the surface was seen in

Figure 6-24; this behavior dropped sharply as surface enamel transitioned into bulk.

Discussion

MI PasteTM Treatments

Upon remineralization with MI PasteTM, AFM imaging suggests that a reaction between

the demineralized enamel and the paste occurred sporadically on the surfaces of both EDTA- and

110




























Figure 6- 1. AFM images of EDTA-treated enamel (30 min) subjected to MI PasteTM for 5 min.
Scale = 20 Crm2


Figure 6- 2. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI PasteTM for
5 min.









buffered lactic acid-treated samples. This is represented by patches of textured enamel not

observed through pure demineralization. Through visual observation it is unclear whether the

surface layer consists of precipitates, enamel that has undergone a phase transformation or a

combination of both. In FE-SEM images of the EDTA-treated substrate, MI PasteTM appears to

have enhanced the sizes and/or accumulation of enamel crystallites in the central and peripheral

regions of the enamel rods. One particular claim as to the functionality of the paste is the ability

to bind calcium and phosphate to dental surfaces.[139] Crystallite agglomeration may have

occurred as a result of the adhesion forces between the paste and apatite. A previously-done

FE-SEM study of remineralization using a diluted CPP-ACP paste reported rather smooth

enamel surface appearance upon subsequent demineralization, barring faint enamel rod

outlines.[140] This supports the theory that MI PasteTM discourages demineralization, perhaps

through the additional buffering action of an actual surface layer that adheres to dental tissue. A

remnant of this layer is perhaps seen on the buffered lactic acid substrate. Structural and

chemical analyses would provide insight into the degree of crystallinity (vs. amorphous nature)

present along the remineralized surfaces, as well as identify any new phases formed.

A representative GIXD spectra comparison of an MI PasteTM-treated sample on an EDTA-

demineralized substrate showed a decrease in crystallinity in the apatite c-axis direction. This can

be attributed to a dominating surface layer that may be partly amorphous in nature, limiting

GIXD detection of crystalline apatite along the (002) plane. A complementary GIXD depth

profile of crystallite size changes from the surface toward the bulk showed the potential for

increased crystallinity (on average) from an as-received enamel state, but the sample size was too

small to make a statistically sound assessment. Of particular interest was the decrease in enamel

crystallinity depicted by the (200) reflection upon EDTA demineralization and the subsequent









enamel surface layers. Selected buffered lactic acid-treated enamel slabs were remineralized with

MI PasteTM and the calcifying solution for microstructural and limited crystallinity comparison

purposes only.

Materials and Methods

Enamel Remineralization Treatments

Selected EDTA-treated samples evaluated in Chapter 5 were subjected to MI PasteTM for

5- and 10-min time intervals as outlined in Chapter 3. Other EDTA-treated samples were

immersed in a calcifying solution for 8 h. For comparison purposes, selected buffered lactic-

acid-treated samples were similarly remineralized. Characterization was performed as outlined in

Chapters 3 and 4. Since limited data were generated for the buffered lactic acid substrate

samples, the statistical significance of those results were not confirmed in this study.

Remineralized Enamel Characterization

Two-dimensional AFM images and representative roughness (RMS) values of the

remineralized enamel samples were generated. Topographical comparisons of the remineralized

surface enamel, as a result of remineralizing agent type, were made.

Secondary electron SEM images of remineralized enamel samples were generated.

Remineralization effects as a result of remineralization treatment type were explored.

High resolution x-ray scattering was performed on remineralized enamel samples. Special

emphasis was placed on discerning the dimensional changes in enamel crystallites upon

demineralization in the c-axis direction.

Nanomechanical studies were performed on remineralized enamel samples. These results

were intended to highlight the structural stability of surface enamel subjected to differing sources

of remineralization.









crystallite size in the c-axis, [002], direction also supports this conclusion. It is interesting note

that redeposition did not follow the crystal orientation pattern observed at the surface, but

showed greater similarity to that seen in bulk enamel. Redeposition of enamel crystallites during

demineralization has been reported, and is described in terms of the self-inhibiting dissolution

behavior of HAP.1126' 127] This idea was based on the concept of constant composition solution

techniques used for enamel remineralization studies.[128] A form of the constant composition

technique called flow-through, or high volume, controls the saturation levels of important ions in

solution necessary for remineralization.[123] Unbalanced ion concentrations in solution do not

occur, which may cause unfavorable shifts in the calcium and phosphate supersaturation levels

required for remineralization. The remineralization process will not cease prematurely. Based on

this concept, it was hypothesized that undersaturation of the treatment solution will encourage

demineralization. Once the solution becomes saturated with HAP structural ions,

demineralization will cease. Intuitively, enamel dissolution effects observed through AFM and

SEM imaging would translate to a decrease in crystallinity of the apatite mineral. These effects

were initially thought to be masked by this occurrence of enamel redeposition/growth. As such,

larger volumes of solution were used to more closely simulate the constant-composition

experimental design.

EDTA- and buffered lactic acid-based experiments produced contradictory results

regarding enamel surface crystallinity. Enamel progressively treated with EDTA showed an

increase in crystallite size in the c-axis direction as a function of treatment time. This is

counterintuitive, as acid attack is known to degrade the mineral structure of enamel. In an

individual crystallite, enamel dissolution begins along the c-axis of the central core region. A

dark line in this region, thought to represent a different form of apatite or a planar defect, was









within the surface layers. This becomes important when choosing ways to treat early signs of

tooth decay in hopes of preventing cavitation.





































Figure 2- 6. Factors contributing to caries lesion development in vivo.[64]


Figure 2- 7. SEM image of enamel eroded with citric acid for 30 min.[20]










Table 5-1. Comparison of as-received and demineralized enamel hardness and modulus values.
Treatment Hardness (GPa) Elastic Modulus (GPa)


As-Received 4.86 + 0.44
As-Received Dentin* 0.59 0.26
0.12M EDTA- 10 min 4.87 + 1.26
0.12M EDTA- 60 min 0.28 0.10
Lactic Acid 37C 1.65 0.36
Lactic Acid 200C 1.74 + 0.87
*Included for comparison purposes only


5

4


95.87 + 5.58
21.08 + 6.44
102.63 + 18.87
4.82 1.25
73.27 + 8.87
63.04 18.08


2 -1---


0 -----


I


As-Received


10 min EDTA


60 min EDTA


Figure 5- 11. Comparison of as-received and EDTA-treated enamel nanohardness values. Error
bars represent 95% confidence.










[121] M. Kuhar, P. Cevc, M. Schara, N. Funduk, JPI,,ithe Dent 1997, 77, 578.

[122] M. D. Francis, NYAcadSci 1965, 131, 694.

[123] J. A. K. M. Buskes, J. Christoffersen, J. Arends, Caries Res 1985, 19, 490.

[124] L. M. Silverstone, British Dental J1968.

[125] S. Hoffman, W. S. McEwan, C. M. Drew, JDent Res 1969, 48, 1234.

[126] L. J. Wang, R. Tang, T. Bonstein, P. Bush, G. H. Nancollas, JDent Res 2006, 85, 359.

[127] S. Mafe, J. A. Manzanares, H. Reiss, J. M. Thomann, P. Gramain, JPhys Chem 1992,
96, 861.

[128] D. J. White, Adv Dent Res 1995, 9, 175.

[129] M. A. Listgarten, Arch Oral Biol 1966, 11, 999.

[130] A. W. Adamson, A. P. Gast, Physical chemistry of surfaces, Wiley-Interscience, New
York 1997.

[131] S. N. Danilchenko, C. Moseke, L. F. Sukhodub, B. Sulkio-Cleff, Cryst Res Technol 2004,
39, 71.

[132] M. S. Bobji, S. K. Biswas, JMater Res 1998, 13, 3227.

[133] R. Z. LeGeros, JDentRes 1990, 69 567.

[134] K. Collys, R. Cleymaet, D. Coomans, D. Slop, JDent 1991, 19, 230.

[135] H. R. Muhlemann, K. Rossinsky, A. Schatt, K. G. Konig, Helv OdontActa 1966, 10, 28.

[136] H. Delvin, M. A. Bassiouny, D. Boston, J Oral Rehabil2006, 33, 26.

[137] A. P. M. Jacobson, R. Strang, K. W. Stephen, Caries Res 1991, 25, 230.

[138] C. D. Gibbs, S. E. Atherton, E. Huntington, R. J. Lunch, R. M. Duckworth, Arch Oral
Biol 1995, 40, 879.

[139] E. C. Reynolds, JDentRes 1997, 76, 1587.

[140] M. Oshiro, K. Yamaguchi, T. Takamizawa, H. Inage, T. Watanabe, A. Irokawa, S. Ando,
M. Miyazaki, J Oral Sci 2007, 49, 115.











-*-As-Received 30 min Room Temp
34

32 Calcifying Soln 8 h
? 32

30

S28

S26

u 24

22

20 .......
0.15 0.2 0.25 0.3 0.4 0.8 1.0 T-2T

Grazing Incidence Angle, w ()


Figure 6- 24. Representative depth profile of the average enamel crystallite size along the c-axis
as a function of co for as-received and 30 min buffered lactic acid-treated enamel,
ending with an 8 h calcifying solution treatment.



































To my family









Table 6-1. Comparison of hardness and elastic modulus values of as-received, EDTA-
demineralized and enamel remineralized with MI PasteTM and a calcifying solution
(0.1 ppm F).
Treatment Hardness (GPa) Elastic Modulus (GPa)
As-Received 4.9 + 0.4 95.9 + 5.6
0.12M EDTA 60 min 0.3 0.1 4.8 1.3
0.12M EDTA + MI PasteTM 0.2 0.2 19.1 + 13.5
0.12M EDTA + Calcifying Soln 0.4 + 0.3 12.6 + 8.1


60 min EDTA EDTA + 5 min EDTA + 8 h
MI PasteTM Calcifying Soln


Figure 6- 11. Comparison of EDTA-treated and subsequently remineralized enamel
nanohardness values. Error bars represent 95% confidence.









inward to the DEJ. For human incisor teeth, a difference in enamel thickness was reported by

Gillings and Buonocore, ] with the greatest thickness at approximately 2 mm from the occlusal

surface of the tooth. The natural variance seen in the tooth, a true biological material, will affect

the treatment and characterization results obtained from each enamel sample.

In this experimental study, as-received human incisor enamel slabs were used to determine

baseline enamel material properties. Surface morphology was examined using AFM and SEM.

Comparisons were made using the conventional 0-20 XRD and GIXD techniques to examine the

crystal structure within the bulk and surface layers of enamel, respectively. The nanoindentation

technique was used to provide nanohardness and elastic moduli data. Confocal Raman data were

extracted to provide a depth profile and chemical data within the enamel surface layers. In

combination, these surface characterization techniques provided an adequate description of

prominent characteristics of tooth enamel affected by demineralization and remineralization.

Materials and Methods

Enamel slabs were prepared and characterized as outlined in Chapter 3.

AFM

Two-dimensional images and representative roughness (RMS) values were taken of

as-received enamel using the AFM. Observation of the surface morphology, without the possible

obstruction of coatings required for SEM analysis, was exploited.

SEM

Secondary electron images were taken of as-received enamel using the FE-SEM. Image

magnification was greater than the sizes of the primary enamel microstructure, namely the

enamel rods.









surface vs. bulk enamel crystallites, Sakae[38] used XRD to analyze powdered enamel samples

from both inner and outer enamel layers. Peak data from the (002) reflection of outer layer

enamel showed an average FWHM value of 0.23, corresponding to an average unit cell length in

the c-direction of 0.6884 nm. The (300) reflection, an indication of crystallinity in the a-axis

direction, showed average FWHM and unit cell length values of 0.3000 and 0.9450 nm,

respectively. Inner layer enamel exhibited slightly larger FWHM and unit cell length values

compared with outer enamel, suggesting a decrease in enamel crystallite size with depth into the

tooth in all directions. While conventional XRD techniques can provide a direct structural

evaluation of tooth enamel, it lacks the surface sensitivity to detect changes in the near-surface

layers.

Transmission Electron Microscopy (TEM)

TEM characterization is based on examining the microstructure of enamel crystallites, as

well as analyzing the lattice periodicity of thin sections of intact enamel. The hexagonal shape of

enamel crystallites has been verified through bright-field imaging studies (Figure 2-14). Of

particular interest is the presence of a dark line bisecting the crystallite thickness in untreated

enamel. White spots were also spotted in the vicinity of the line[39]. Marshall and Lawless440]

observed a similar phenomenon, suggesting that the line represents a planar defect along the

(100) plane. Expounding upon this theory is the suggestion of a localized modification in

chemical composition, perhaps indicating the presence of a crystalline phase. One suggestion is

that this core region contains carbonate, an impurity known to make HAP more susceptible to

chemical dissolution. There may be some validity to this theory in that the core regions of

enamel crystallites are often the first to show signs of acid dissolution via TEM (Figure 2-15).

Simmelink and Abrigo[41] reported centralized crystallite dissolution processes with duration of









closely mimics the changes in pH experienced in vivo due to eating and drinking throughout the

day.

The phenomenon of remineralization, as well as the influence of fluoride, has been

extensively reported in the literature. Amaechi et al. [52] reported an increase in mineral content

after al-hr exposure to orange juice in situ during subsequent exposure to the participant's own

saliva. MR detected at least a 48% decrease in lesion depth due to remineralization. Silverstone

and Wefel531 reported an increase in enamel mineral content, driven by an increase in crystallite

diameter, when subjected to calcifying remineralization solutions. Collys etal. [54] reported an

increase in enamel microhardness as a result of remineralization via calcifying solution in the

presence of low levels of fluoride or enhanced levels of phosphate. Conversely, Lippert et al. 33]

reported no significant increase in softened enamel protection from toothbrush abrasion as a

result of artificial saliva remineralization analyzed via nanoindentation studies, suggesting the

formation of an unstable remineralized enamel phase. The same group of researchers further

substantiated their claim by conducting pH cycling studies and subsequent nanoindentation tests

on the treated enamel. A collapse of the enamel surface layer due to subsurface demineralization

was reported during testing, yielding results largely unrepresentative of the true nature of the

remineralized layer. 551 Aoba et al. [56] observed a decrease in FWHM and an increase in the

integrated intensities of the (004) and (310) reflections along the surface of artificial HAP pellets

upon acid dissolution. A well-mineralized surface layer formed on the pellets during subsurface

dissolution, an occurrence that some researchers label as "remineralization" of the HAP surface

layers. These results suggest an improvement in HAP crystallinity along the surface, driven by

an overall increase in crystallite size. A study performed by Vieria et al. [57 used finely-ground

human enamel samples subjected to fluoridated water from various regions for 0-20 XRD









[82] C. D. Gibbs, S. E. Atherton, E. Huntington, R. J. M. Lynch, R. M. Duckworth, Arch Oral
Biol 1995, 40, 879.

[83] J. L. Linton, Am J OrthodDentofacial Orthop 1996, 110, 590.

[84] J. B. D. Featherstone, Caries Res 1981, 15, 377.

[85] B. T. Amaechi, S. M. Higham, JDent 2001, 29, 371.

[86] E. Saxegaard, F. Lagerlof, G. Rolla, Acta Odontol Scand 1988, 46, 355.

[87] S. V. Dorozhkin, JCollInt Sci 1997, 191, 489.

[88] H. U. V. Gerth, T. Dammaschke, E. Schafer, H. Zuchner, DentMater 2007, 23, 1521.

[89] E. Saxegaard, G. Rolla, ScandJDent Res 1988, 96, 523.

[90] K. Yamaguchi, M. Miyazaki, T. Takamizawa, H. Inage, B. K. Moore, JDent 2006, 34,
230.

[91] D. Tantbirojn, A. Huang, M. D. Ericson, S. Poolthong, JDent 2008, 36, 74.

[92] P. Shen, F. Cai, A. Nowicki, J. Vincent, E. C. Reynolds, JDent Res 2001, 80.

[93] G. Walker, F. Kai, P. Shen, C. Reynolds, B. Ward, C. Fone, JDairy Res 2006, 73, 74.

[94] M. I. Richard, T. U. Schulli, E. Wintersberger, G. Renaud, G. Bauer, Thin SolidFilms
2006, 508, 213.

[95] J. F. Moulin, F. Dinelli, M. Massi, C. Albonetti, R. Kshirsagar, F. Biscarini, Nucl.
Instrum. Methods Phys. Res., Sect. B 2006, 246, 122.

[96] K.-J. Chao, P.-H. Liu, K.-Y. Huang, C. R. Chimie 2005, 8, 727.

[97] D. C. Creagh, V. Otieno-Alego, P. M. O'Neill, Radiat. Phys. Chem. 2001, 61, 757.

[98] I.-M. Low, JAm Ceram Soc 2004, 87, 2125.

[99] C. H. Russell, "X-ray reflectivity study of exotic materials for electronic materials",
presented at Materials Research Society Symposium, 2002.

[100] J. Daillant, A. Gibaud, Eds., X-ray andNeutron Reflectivity: Principles and Applications,
Springer-Verlag, Berlin 1999.

[101] H. Toraya, JAppl Cryst 1986, 19, 440.









Materials and Methods

Enamel Demineralization Treatments

Select samples evaluated in Chapter 4 were immersed in 0.12M EDTA for 10, 30 and 60

min time intervals as outlined in Chapter 3. The remaining as-received samples were immersed

in a buffered lactic acid solution for 30 min at human body (370C) and room (200C)

temperatures. The treated samples were characterized as outlined in Chapter 3, unless otherwise

indicated.

Demineralized Enamel Characterization

Two-dimensional AFM images and representative roughness (RMS) values of enamel

treated with EDTA and buffered lactic acid were generated. Amplitude signal images,

representative of the topographical changes experienced by samples, were produced.

Additionally, height signal images of the same regions mapped surface depth information. A

flattening function was performed on all height-signaling images to remove unwanted artifacts

prior to roughness analyses.

Additional SEM images of enamel treated with EDTA were generated. Demineralization

patterns in surface enamel as a result of acid treatment time were explored.

High resolution x-ray scattering was performed on enamel treated with EDTA and buffered

lactic acid. In some cases, conventional 0-20 XRD scans were generated for comparison

purposes. Special emphasis was placed on evaluating the dimensional changes in enamel

crystallites upon demineralization in the c-axis direction.

Nanomechanical studies were performed on enamel treated with EDTA and buffered lactic

acid as outlined in Chapter 4. These results were intended to highlight the structural stability of

surface enamel subjected to varying degrees of demineralization.









a specified lateral position is calculated using a formula developed by Angmar et al.[24] (Equation

2-2), where An,slice(x) is the x-ray absorbance of the tooth at a specified lateral

V(x) = 100 (An,sice(x) / t) io ) / (lm go) (vol %) (2-2)

position, t is the thickness of the tooth sample, Cm is the linear x-ray attenuation coefficient of

HAP, go is the attenuation coefficient of the organic component of the tooth. A typical MR plot

modeling a subsurface caries lesion is shown in Figure 2-10. This concept has been expounded

upon and extended to clinical use in the form of dental x-rays. This technique serves as a great

aid to visual and manual evaluations of dissolved enamel.

As caries lesions initiate and progress beneath the tooth surface, this method has proven to

be an effective means of identifying this form of tooth decay. A major limitation of the TMR

technique is detection resolution. TMR is unable to observe enamel mineral changes less than

10 .m from the tooth surface due to densitometer slit restrictions and sample curvature.[25] This

poses a problem when attempting to evaluate the earliest stages of caries development and

erosion. At best, MR can indirectly measure structural changes in demineralized enamel on the

micron scale.

Profilometry

Profilometry, or surfometry, is a method typically used to measure erosion along the

enamel surface. 23, 26-28] Traditional experiments have used a diamond stylus, tens of microns in

diameter, to scan across the sample surface at a specified rate. A comparison of profilometric

traces produced by both untreated and eroded sample surface scans can provide a lesion depth

profile with micron resolution. Profilometric experiments conducted with a laser stylus provide

the added benefit of a potential three-dimensional interpretation of the erosion lesion due to

focusing and maneuverability of the laser beam, an example of which is shown in Figure 2-11.

The laser beam does not physically touch the sample surface, eliminating potential damage






























*1


m-I ED1 Sample D 43 hys B
3 02-4 0201 EDTA200431rh
1600l- -j -- --- --- ---- --- --- f --- -f
150- --- -- --- --- --___ --- -L
1600-
1500-



114



900-








200-- -_ __- I
W1o -- -- -- --- -- -- -- / -- -- -- -- f -


.O)0 500 0 pa C' 3:.', 4000
DipLacpumer (nm)


450 0 50C,


Figure 5- 14. Comparison of representative force (load)-displacement curves for EDTA-treated
enamel and as-received dentin. A) 60 min EDTA treatment. B) As-received dentin.


Figure 5- 15. Confocal microscopy images of as-received enamel treated with 0.12M EDTA for
60 min. A) X-Y surface scan. B) X-Z cross-sectional scan.


0 00 -0 1 o,0 156C.









BIOGRAPHICAL SKETCH

Carmen Veronica Gaines was born in Buffalo, New York. She obtained her secondary

educational training from the Hutchinson Central Technical High School, with a concentration in

computer technology. Upon graduation, Carmen went on to pursue a bachelor of science degree

at Cornell University, majoring in biological engineering. She was later able to put her problem

solving skills to work by joining the team at Harmac Medical Products, Inc. as a product

development engineer. Her interest in the biomedical field was solidified there, and encouraged

the pursuit of more formalized education and an opportunity to conduct research at the

University of Florida. She obtained her Master's degree from the Department of Materials

Science and Engineering in May 2005 and completed her Ph.D. from the same department in

May 2008. Carmen hopes to continue her research pursuits in the area of artificial organ

development.




Full Text

PAGE 1

1 NEAR-SURFACE STRUCTURAL EXAMINATION OF HUMAN TOOTH ENAMEL SUBJECT TO IN VITRO DEMINERALIZATION AND REMINERALIZATION By CARMEN VERONICA GAINES A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Carmen Veronica Gaines

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGMENTS First, I would like to thank God for His blessings and never-ending presence in my life. I would like to thank my family and friends for the endless support, encouragement and love provided throughout the years, and especially throughout the dissertation process. I would also like to thank my dissertation chair, Dr. Vale ntin Craciun, as well as my committee membersDr. Amelia Dempere, Dr. Ke nneth Anusavice, Dr. Susan Sinnott and Dr. Yasumasa Takano. They have all enriched my gr aduate school experien ce in various ways from engaged discussions of the research, fi nancial assistance for business travel and conferences, general encouragement during rough pa tches to agreeing to serve as members of my supervisory committee. Special thanks go to Mr. Ben Lee and Mr. Nai-Zheng Zhang for assistance with sample and experimental solution preparation. The two of you always made me feel welcome during the time I spen t in the Dental labs. I would al so like to thank Drs. Luis and Barbara Muga for kindly steriliz ing teeth for my experiments a nd for the great conversations. Data obtained for my research would not have been possible w ithout the help of the MAIC staffthank you all. I would also like to expre ss appreciation for my labmates, Chuchai and Junghun, for their encouragement and he lp with sample characterization. Lastly, I would like to thank the UF Material s Science and Engineering department for the opportunity to pursue my research and providi ng opportunities for financial assistance.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4LIST OF TABLES................................................................................................................. ..........8LIST OF FIGURES.........................................................................................................................9ABSTRACT...................................................................................................................................14 CHAP TER 1 INTRODUCTION..................................................................................................................16Prevalence of Chemical Tooth Decay.................................................................................... 16Clinical Concerns.............................................................................................................. ......17Outline of Dissertation........................................................................................................ ....182 LITERATURE REVIEW.......................................................................................................20Tooth Structure.......................................................................................................................20Enamel Material Properties.................................................................................................... 21Enamel Demineralization....................................................................................................... 22Microradiography (MR)..................................................................................................24Profilometry................................................................................................................... ..25Mechanical Testing......................................................................................................... 26X-ray Diffraction (XRD)................................................................................................. 27Transmission Electron Microscopy (TEM)..................................................................... 29Enamel Remineralization........................................................................................................ 303 EXPERIMENTAL METHODS.............................................................................................44Enamel Sample Preparation....................................................................................................44Enamel Surface Treatments.................................................................................................... 45Demineralization.............................................................................................................450.12M Ethylenediamine tetraacetic acid (EDTA)....................................................45Buffered lactic acid.................................................................................................. 46Remineralization..............................................................................................................47Calcifying solution................................................................................................... 47MI Paste topical cream......................................................................................... 47Enamel Characterization.........................................................................................................48High Resolution X-ray Scattering................................................................................... 49X-ray reflectivity (XRR)..........................................................................................50Grazing incidence x-ray diffraction (GIXD)............................................................ 52Experimental set-up.................................................................................................. 52Nanoindentation..............................................................................................................53

PAGE 6

6 Confocal Raman Microscopy.......................................................................................... 54Scanning Electron Microscopy (SEM)............................................................................54Atomic Force Microscopy (AFM)................................................................................... 55Attempted TEM Characterization................................................................................... 55Statistical Considerations................................................................................................564 ENAMEL CHARACTERIZATION...................................................................................... 62Introduction................................................................................................................... ..........62Materials and Methods...........................................................................................................63AFM................................................................................................................................63SEM.................................................................................................................................63XRD.................................................................................................................................64Nanoindentation..............................................................................................................64Confocal Raman Microscopy.......................................................................................... 65Results.....................................................................................................................................65Discussion...............................................................................................................................66Tooth Enamel Defined....................................................................................................66Characterization Concerns...............................................................................................675 ENAMEL DEMINERALIZATION....................................................................................... 75Introduction................................................................................................................... ..........75Materials and Methods...........................................................................................................77Enamel Demineralization Treatments.............................................................................77Demineralized Enamel Characterization......................................................................... 77Results.....................................................................................................................................78Demineralization 0.12M EDTA.................................................................................. 78Demineralization Buffered Lactic Acid...................................................................... 81Discussion...............................................................................................................................826 ENAMEL REMINERALIZATION..................................................................................... 104Introduction................................................................................................................... ........104Materials and Methods.........................................................................................................106Enamel Remineralization Treatments........................................................................... 106Remineralized Enamel Characterization....................................................................... 106Results...................................................................................................................................107MI Paste Treatments.................................................................................................. 107EDTA-treated enamel substrate............................................................................. 107Buffered lactic acid-treated enamel substrate........................................................ 109Calcifying Solution Treatments..................................................................................... 109EDTA-treated enamel substrate............................................................................. 109Buffered lactic acid-treated enamel substrate........................................................ 110Discussion.............................................................................................................................110MI Paste Treatments.................................................................................................. 110Calcifying Solution Treatments..................................................................................... 114

PAGE 7

7 7 CONCLUSIONS.................................................................................................................. 129Important Findings from the Research................................................................................. 129Future Work..........................................................................................................................131 APPENDIX A ENAMEL POLISHING........................................................................................................ 133LIST OF REFERENCES.............................................................................................................136BIOGRAPHICAL SKETCH.......................................................................................................144

PAGE 8

8 LIST OF TABLES Table page 21. Minor inorganic constituents in sound hum an tooth enamel................................................. 3722. Sampling of sound tooth enamel hardness and modulus values........................................... 413-1. Composition of buffered lactic acid demineralization solution.............................................. 583-2. Composition of calcifying remineralization solution............................................................. 583-3. Raman frequencies and PO4 3vibrational modes associated with tooth enamel.................... 615-1. Comparison of as-received and demineraliz ed enamel hardness and modulus values.......... 946-1. Comparison of hardness and elastic modulus values of as-received, EDTAdemineralized and enamel remineralized w ith MI Paste and a calciyfing solution (0.1 ppm F).................................................................................................................... ...121

PAGE 9

9 LIST OF FIGURES Figure page 21. Schematic of an enamel crystallite....................................................................................... .3522. Schematic of enamel rods, highli ghting crystallite and rod orientation................................3523. Photograph of a human tooth slice showing enamel rod orientation.................................... 3624. Schematic highlighting ma jor components of the tooth........................................................ 3625. The idealized crystal structure of HAP, as viewed along the c-axis.....................................3726. Factors contributing to caries lesion development in vivo ....................................................3827. SEM image of enamel eroded with citric acid for 30 min..................................................... 3828. Photograph of facial erosion with shiny and smooth appearance......................................... 3929. Scanning electron microscope of enamel etched with EDTA...............................................39210. Schematic of an average en amel mineral content MR profile............................................. 40211. Profilometric trace of eroded enamel lesion by a demi neralizing agent for 30 min........... 40212. Typical load-displacement curve produced by nanoindentation......................................... 41213. XRD schematic illu strating Braggs Law............................................................................42214. TEM image of a cross-section of an enamel crystallite c-axis............................................ 42215. TEM image of cross-sections of demineralized enamel crystallite caxes......................... 43216. Schematic illustrating the balance of demineralization and remineralization..................... 4331. Extracted human tooth and epoxied enamel slab.................................................................. 5732. Solubility isotherms expressing the relati onship between the enamel mineral ions (Ca) and pH at saturation for HAP and FAP............................................................................. 5733. Schematic illustrating the analogy of an enamel surface layer on an untreated, sound enamel substrate to a thin film model................................................................................ 5834. PANalytical XPERT MRD PRO four-axis set-up..............................................................5935. GIXD geometry.....................................................................................................................60

PAGE 10

10 36. Comparison of XRD spectra of TiO2 on a glass substrate produced by conventional -2 geometry and GIXD surface-specific geometry........................................................60 37. X-ray attenuati on length of Cu K radiation in HAP............................................................ 6141. Average values of enamel hardness and elastic modulus...................................................... 6942. AFM images of as-received human tooth enamel................................................................. 7043. FE-SEM image of as-received enamel.................................................................................. 7044. Conventional -2 XRD comparison of as-received enamel and NIST calcium HAP standard #2910...................................................................................................................71 45. Comparison of as-received en am el analyzed by conventional -2 XRD and GIXD ( = 0.4). ............................................................................................................... 71 46. Representative XRR curve of as-received enamel................................................................ 7247. ProFit (002) peak modeling............................................................................................... ....7248. Depth profile of the average crystallite sizes along the c-axis from the enamel surface into the bulk.......................................................................................................................7349. Representative force (load)-displacement nanoindentation curve for as-received enamel......................................................................................................................... .......73410. Confocal microscopy images of as-received enamel.......................................................... 74411. Averaged Raman spectrum of as-received enamel. .............................................................7451. AFM images highlighting transfor mation of as-received enamel upon 0.12M EDTA demineralization for 30 min............................................................................................... 89 52. FE-SEM image highlighting transfor m ation of as-received enamel upon 0.12M EDTA demineralization for 30 min.............................................................................................. 89 53. FE-SEM image of as-received enamel treated with 0.12M EDTA for 10 m in..................... 9054. FE-SEM image of as-received enamel treated with 0.12M EDTA for 60 min..................... 9055. Representative GIXD spectra comparison of as-received enamel treated with < 25 ml of 0.12M EDTA solution at = 0.8.................................................................................91 56. Representative GIXD spectra comparison of as-receiv ed and EDTA-treated enamel at = 0.4..............................................................................................................................91

PAGE 11

11 57. Depth profile of the averag e en amel crystallite size alon g the c-axis as a function of for EDTA enamel treatment regime.................................................................................. 9258. Depth profile of the average surface enam el crystallite sizes along the c-axis as a function of for EDTA enamel treatment regime............................................................ 9259. Correlation of EDTA treatment time with crystallite size at = 0.15................................93510. Representative XRR curve comparison of as-received and EDTA-treated (30 min) enamel................................................................................................................................93511. Comparison of as-received and EDTA -treated enamel nanohardness values..................... 94512. Comparison of as-received and EDTA-t reated enamel elastic modulus values.................. 95513. Comparison of representative force (loa d)-displacement curves for as-received and EDTA-treated enamel........................................................................................................ 95514. Comparison of representative force (l oad)-displacement curves for EDTA-treated enamel and as-received dentin...........................................................................................96515. Confocal microscopy images of as-receiv ed enamel treated with 0.12M EDTA for 60 min................................................................................................................................96516. Averaged Raman spectrum of as-recei ved enamel treated with 0.12M EDTA for 60 min ...............................................................................................................................97 517. AFM images of as-received enamel treat ed with buffered lactic acid for 30 m in at room temperature............................................................................................................... 97518. AFM images of as-received enamel treat ed with buffered lactic acid for 30 min at human body temperature.................................................................................................... 98519. SEM images of as-received enamel treat ed with buffered lactic acid for 30 min at room temperature............................................................................................................... 98520. Depth profile of the averag e enamel crystallite size alon g the c-axis as a function of for a 30 min buffered lactic acid treatment at human body temperature........................... 99521. Representative GIXD spectra comparison of as-received and buffered lactic acidtreated enamel at = 0.4..................................................................................................99522. Depth profile of the averag e enamel crystallite size alon g the c-axis as a function of for a 30 min buffered lactic acid treatment at room temperature.................................... 100 523. Representative XRR curve comparison of as-received enamel and treated with buffered lactic acid for 30 min at room temperature....................................................... 100

PAGE 12

12 524. Comparison of as-received and buffered lactic acid-treated enam el nanohardness values...............................................................................................................................101525. Comparison of as-received and EDTA-t reated enamel elastic modulus values................ 101526. Comparison of representative force (loa d)-displacement curves for as-received and 30 min buffered lactic acid-treated enamel...................................................................... 102527. Confocal microscopy X-Y surface image of as-received enamel treated with buffered lactic acid for 30 min at room temperature...................................................................... 102528. Averaged Raman spectrum of enamel tr eated with buffered lactic acid for 30 min at room temperature. ........................................................................................................103 529. TEM image showing the bonding of neighbori ng crystallites within a caries lesion. ....... 10361. AFM images of EDTA-treated enamel ( 30 min) subjected to MI Paste for 5 min........ 11662. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI Paste for 5 min.....11663. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI Paste for 5 min, highlighting a single enam el rod......................................................................................11764. FE-SEM image of EDTA-treated enamel (30 min) subjected to MI Paste for 5 min, highlighting central region of a single enam el rod.......................................................... 11765. Representative GIXD spectra comparis on of as-received, ED TAand MI Pastetreated enamel at = 0.4................................................................................................11866. (002) GIXD spectra comparison for as -received, EDTA-treated (30 min) and MI Paste-treated (5 and 10 min) enamel at = 0.4. .................................................. 118 67. Depth profile of the averag e en amel crystallite size alon g the c-axis as a function of for as-received enamel progressively tr eated with EDTA, en ding with a 5 min MI Paste treatment....................................................................................................... 119 68. Depth profile of the average surface enam el crystallite sizes along th e c-axis as a function of for as-received enamel progressively treated with ED TA, ending with a 5 min MI Paste treatment............................................................................................. 11969. (002) GIXD spectra comparison for as -received, EDTA-treated (30 min) and MI Paste-treated (5 and 10 min) enamel at = 0.4. .................................................. 120

PAGE 13

13 610. Representative depth profile of the averag e enam el crystallite size along the c-axis as a function of for as-received enamel treated w ith EDTA for 60 min, ending with a 5 min MI Paste treatment............................................................................................. 120611. Comparison of EDTA-treated and subsequently remineralized enamel nanohardness values. ..............................................................................................................................121612. Comparison of EDTA-treated and subsequently remineralized enamel elastic modulus values.. .............................................................................................................................122613. Comparison of representative force (l oad)-displacement curv es for as-received, 60 min EDTA-treated and remineralized enamel............................................................ 122 614. X-Y surface images of 60 min EDTA + M I Paste-treated (5 min) enamel................... 123615. Confocal microscope X-Z cross-sec tional image of 60 min EDTA + MI Pastetreated (5 min) enamel.....................................................................................................123616. Averaged Raman spectrum of 60 min EDTA + MI Paste-treated (5 min) enamel....... 124617. AFM images of buffered lactic acid-treat ed enamel (room temperature) subjected to MI Paste for 5 min.......................................................................................................124618. Representative depth profile of the averag e enamel crystallite size along the c-axis as a function of for as-received and buffered lactic acid-treated enamel, ending with a 5 min MI Paste treatment............................................................................................. 125619. AFM images of 60 min ED TA-treated enamel subjected to a calcifying solution for 8 h.....................................................................................................................................125620. GIXD spectra comparison of as-receiv ed, EDTAand calcifying solution-treated enamel at = 0.4............................................................................................................ 126621. Representative depth profile of the averag e enamel crystallite size along the c-axis as a function of for as-received and 60 min EDTA-treat ed enamel, ending with a 5 min MI Paste and a 8 h cal cifying solution treatment..............................................126622. AFM images of buffered lactic acid-treated enamel (room temperature) subjected to a calcifying solution for 8 h................................................................................................127623. FE-SEM image of 30 min buffered lactic acid-treated enamel (room temperature) subjected to a calcifying solution for 8 h, highlighting the central region of a single enamel rod..................................................................................................................... ...127624. Representative depth profile of the averag e enamel crystallite size along the c-axis as a function of for as-received and 30 min buffe red lactic acid-treated enamel, ending with an 8 h calcifying solution treatment............................................................. 128

PAGE 14

14 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy NEAR-SURFACE STRUCTURAL EXAMINATION OF HUMAN TOOTH ENAMEL SUBJECT TO IN VITRO DEMINERALIZATION AND REMINERALIZATION By Carmen Veronica Gaines August 2008 Chair: Valentin Craciun Major: Materials Science and Engineering The early stages of chemical tooth decay are governed by dynamic processes of demineralization and remineralizati on of dental enamel that init iates along the surface of the tooth. Conventional diagnostic tech niques lack the spatial resoluti on required to analyze nearsurface structural changes in enamel at the s ubmicron level. In this study, slabs of highlypolished, decay-free human enamel were subjec ted to 0.12M EDTA and buffered lactic acid demineralizing agents and MI Paste and cal cifying (0.1 ppm F) remineralizing treatments in vitro Grazing incidence x-ray diffr action (GIXD), a technique typi cally used for thin film analysis, provided depth profile s of crystallinity changes in surface enamel with a resolution better than 100 nm. In conjunction with nanoinde ntation, a technique gaining acceptance as a means of examining the mechanical properties of sound enamel, these results were corroborated with well-established microscopy and Ra man techniques to assess the nanohardness, morphologies and chemical nature of treated enamel. Interestingly, the average crystallite size of surface enamel along its c-axis dimension increased by nearly 40% after a 60 min EDTA treatment as detected by GIXD. This result was in direct contrast to the obvious surface degrada tion observed by microscopi c and confocal Raman imaging. A decrease in nanohardness from 4.86 0.44 GPa to 0.28 0.10 GPa was observed.

PAGE 15

15 Collective results suggest that mineral dissoluti on characteristics evid ent on the micron scale may not be fully translated to the nanoscale in assessing the integrity of chemically-modified tooth enamel. While an intuitive decrease in en amel crystallinity was observed with buffered lactic acid-treated samples, demineralization wa s too slow to adequately quantify the enamel property changes seen. MI Paste treatment of EDTA-demineralized enamel showed preferential growth along the a-axis direction. Calcifying solution treatments of both demineralized sample types appeared to have negligible effects on enamel crystallinity. Both remineralizing agents provided an increase in re siliency within the enamel surface layers. Findings from this study may prove useful in identifying more effective methods to prevent enamel demineralization and to prom ote and/or enhance remineralization for the treatment of tooth decay. Careful considerati on of the nanoscale properties of treated surface enamel may lead to an understa nding of how to truly regenerate decomposed enamel mineral from the inside out.

PAGE 16

16 CHAPTER 1 INTRODUCTION Biological apatite is a major constituent of a variety of naturally-o ccurring hard tissues, namely bone and teeth. Apatite dissolution is know n to compromise the structural integrity and, consequently, the desired functionality of thes e materials. This research examines the degradation of the hardest tissue in the human bodytooth enamelas it re lates to the process of decay, or demineralization. Prevalence of Chemical Tooth Decay Enamel demineralization, and subsequent increas e in surface porosity, is characteristic of the chemical decay and erosion often experienced by the tooth. The formation of acid in dental plaque through the metabolis m of sugary substances in vivo slowly decomposes the enamel surface of the tooth. Left untreated, tooth decay even tually spreads to underlying layers of dental tissue, forming highly-demineralized zones ca lled caries lesions. A likely consequence of advanced tooth decay is irreversible cavitation. In extreme cases, necrosis of the innervated pulpal tissue will occur. The inconvenient, cos tly, and often painful, circumstances surrounding tooth decay may require treatments ranging from protective sealant applic ations to root canal surgeries. Considered the most commonly treated human oral dise ase to date, the battle with tooth decay is likely to be f ought throughout an individuals lifetime. According to the U.S. Department of Health and Human Services, near ly 20% of small children aged 2-4 years have shown signs of potential cavit ation through the development of precursory caries lesions.[1] A vast 80% of the American populat ion has had at least one cavity by late adolescence. Over 66% of adults aged 35-44 years have lost one pe rmanent tooth through car ies-induced decay, and approximately 25% of adults between the ages of 65 and 74 have lost all of their permanent teeth.[1]

PAGE 17

17 Similarly, frequent or prolonged exposure to acidic agents that work to induce enamel erosion is a recurring problem. Data collected from San Antonio, TX, as part of an on-going nationwide study, suggested a 30% prevalence rate among preteens.[2] With detected erosion rates of nearly 50% in Western European c ountries among the teenage and younger populations, erosion is gaining attention in the area of dental pathology.[3] Clinical Concerns Although the advent of fluoridated drinking water and topical fluoride treatments has tremendously improved the state of oral health wo rldwide, the statistics s uggest that more work is needed. The depressed state of oral health in America has prompted a shift in focus from reactive to preventative dental care.[4] The earlier that enamel dissolution can be detected and treated by cost-effective, minimally-invasive mean s, the greater the chance for preservation of the structural integrity and functionality of the to oth. This proposition is es pecially relevant for residents in low socioeconomic environments, who often have limited access to the resources necessary to combat the advanced stages of tooth decay. Progress in early caries treatments, materials selection for tooth re storation and an increased understanding of the overall caries development process has made preventive dental care a possibility.[5] Interestingly, reports have indicated that clinicians are hesitant to adopt this strategy in practice. There is weak evidence supporting the ability of current di agnostic techniques to ac curately detect the earliest stages of tooth decay. A review performed for the 2001 NIH Consensus Development Conference on caries management suggested that several diagnostic testsincl uding microradiographic, laser fluorescence and electrical conductance methodsyiel ded results that were largely inconclusive, with great variance due to operato r error and data relevance. As such, a baseline protocol to clinically assess early stages of t ooth decay has yet to be developed.[6] An increased understanding of the crystallinity, chemical and mechanical changes resulting from early enamel

PAGE 18

18 decomposition correlated with morphological ch anges along the tooth surface layers should prove beneficial. The process of enamel reminera lization, primarily driven by the use of fluoridecontaining topical treatments, re fers to the reconstruction of partially-decomposed enamel. There is much controversy over the underlying m echanisms surrounding this process. Debate most often ensues among the theories of actual enamel mineral regeneration, the formation of a mineral-like substance in place of the decom posed enamel, or enamel strengthening through means other than mineral enhancement. Analysis of the submicron-level changes of the enamel mineral under both demineralization and remine ralization conditions may help to explain crystallinity effects of these processes within surface enamel. Such an understanding from the smallest structural unit of enamel to the macroscal e may prove beneficial in serving as an aid in the future development of preventative treatments. Outline of Dissertation This research attempts to address the afor ementioned concerns posed by the scientific community regarding the earliest stages of tooth decay, with a focus on activity within the enamel surface layers. Chapter 2 describes the overall structure of the tooth and material properties specific to enamel. The dynamic pro cesses of enamel dissolution and regeneration namely demineralization and remineralizati onare explained. A survey of current characterization techniques used to assess cha nges in enamel structural integritysuch as microradiography, profilometry, microhardness testing and conventional x-ray diffraction studieswas performed. The tooth decay process is governed by enamel-oral fluid interactions at the atomic level. As such, current diagnos tic techniques lack the resolution necessary to quantify changes in the surface en amel crystal structure due to early demineralization. The characterization of chemically-modified enamel on the submicron scale is needed. Chapter 3 outlines the general experimental plan for this research, including enamel surface sample

PAGE 19

19 preparation and relevance of the demineraliz ation and remineralization agents used. Highresolution x-ray methods, grazing incidence x-ra y diffraction (GIXD) and x-ray reflectivity (XRR), were used to test the hypoth esis that crystallinity changes can be resolved in near-surface enamel. Nanoindentation, a technique gaini ng acceptance as a means of examining the mechanical properties of sound enamel, was performed. These results were corroborated with well-established microscopy, na nomechanical and Raman techni ques for validation purposes. Chapter 4 provides a thorough characterization of untreated tooth enamel, which served as a baseline for subsequent studies in this research. Chapter 5 examines the effects of enamel demineralization on the structure of surface enamel. Chapter 6 test s the hypothesis that epitaxial regrowth of demineralized enamel occurred due to remineralization, as detected by GIXD. Finally, Chapter 7 provides a summary of all major results, suggested future work and discussion of the implications of the research for the ongoing development of relevant dental decay treatments.

PAGE 20

20 CHAPTER 2 LITERATURE REVIEW Tooth Structure Dental enamel is the hard, protective covering of the crown of the tooth. It is the most highly mineralized of human tissues, surpassing bone. Apatite, a crystalline calcium phosphate, comprises 96 wt% of its mineral content. St ructurally similar to hydroxyapatite (HAP), Ca10(PO4)6(OH)2, enamel crystallites are hexagonal in shape and el ongated along their c-axes (Figure 2-1). Bundles of crystallites form rows of densely-packed rods positioned perpendicular to the tooths surface (Figure 2-2). The crystallites are oriented para llel to the vertical axis of the enamel rod, and fan out to angles of 10-40 along the periphery.[7] The enamel rods measure 4 to 7 m in diameter, and are layere d to give an overall enamel thickness of 1 to 2 mm (Figure 2-3).[8] The remaining 4 wt% of enamel consists of a proteinand polysaccharide-based matter intertwined in a fine network among th e rods and between crystallites, and water.[9] Enamel may be considered a composite material, consisting of ordered arrays of fiber-like crystals held together by a permeable organic matrix. Beneath the enamel layer is the softer, mo re porous, yellowish dentin. Like enamel, it consists of a HAP-like mineral phase comprised of smaller crystallites. This tissue is most similar in structure to bone. Dentin contains only 68 wt% apatite. An organic matrix of collagenous proteins surrounds the mineral, servi ng as the structural backbone of this phase.[10] These microstructural characteristics render dentin more reactive and susceptible to chemical attack relative to enamel. Dentin, the most abunda nt dental tissue, determines the size and shape of the tooth (Figure 2-3). The dentinoenamel j unction (DEJ) serves as an interface between the dentin and enamel. The innervated pulpal cavity is found at the center of the tooth. The root of the tooth is protected by an ex tension of enamel, the cementum, within the oral gingiva.

PAGE 21

21 Enamel Material Properties Stoichiometric HAP most commonly forms a hexagonal crystal structure of the space group P63/m (No. 176), with 3-fold axis symmetry. Its lattice parameters are defined as a = b = 9.432 and c = 6.881 .[11] Calcium, phosphorus and hydroxyl positioning in the HAP unit cell is illustrated in Figure 2-5. Enamels hard, brittl e nature is susceptible to a chipping type of fracture. A combination of the in terlocking pattern of the enamel rods and the resilient dentin sublayer helps to resist crack prop agation and fracture through the DEJ.[10] The translucent coloring of the tooth ranges in shade from light yellow to grayish white. The color is representative of that of dentin, fi ltered by the thickness of the enamel.[12] Enamel has a refractive index of 1.62 and an averag e density in the range of 2.95.97 g/cm3.[10] During enamel formation and maturation, impurity ions penetrate the apat ite crystal lattice, introducing structural defects. The organic matrix provides a di ffusion pathway for travel to deeper tissue layers. Carbonated HAP (cHAP), Ca10(PO4)6(CO3)x(OH)2, 0
PAGE 22

22 Considering the high mineral content of enam el, evaluation of crys tallinity may provide valuable information to better understand the structural implications of early enamel dissolution and remineralization. An important aspect is the examination of the chemical reactions involved in these processes. Sinc e these reactions occur in vivo at the tooth-oral fluid interface, surface enamel should be investigated. Enamel Demineralization In vivo demineralization occurs thro ugh the chemical attack of tooth enamel, resulting in the loss of calcium and phosphate ions (as well as other substitutional impurit ies present, such as carbonate) through dissolution and diffusion. Under ideal conditions sali va is supersaturated with respect to enamel components, containing natural levels of calcium and phosphate at neutral pH. This encourages homeostasis within the mouth.[12] Eating and drinking te nds to upset this balance. In caries lesion developm ent, oral bacteriaspecifically Streptococcus mutans ( S. mutans ) and lactobacilliconvert dietary sugars into weak acids (e.g. lactic, acetic). The bacteria, sugar, and saliva combine to form a biofilm of plaque, which can adhere to tooth surfaces within 20 min of eating.[9] If not removed on a routine basis, the acidic plaque begins to decompose the enamel surface. Acid production cause s the plaque pH to fall below the critical value of approximately 5.5, favoring enamel dissolu tion (Figure 2-6). Basic solubility laws allow for diffusion of the Ca2+, PO4 3-, and OHions from the apatite crysta ls, through the plaque to the saliva, compromising the strength and structur e of enamel. This process is defined by Equation 2-1. Acid dissolution occurs because th e enamel becomes undersaturated with respect to its most Ca10(PO4)6(OH)2 + 8H+ 10Ca2+ + 6HPO4 2+ 2H2O (2-1) important structural constituents due to the removal of these critical ions.[15]

PAGE 23

23 As stated earlier, enamel dissolution through acidic by-products of bacterial origin may give rise to caries lesions. C overing an area that is 20 to 50 m thick, they are found beneath a well-mineralized, protective su rface layer of highly-fluorid ated enamel. The lesions are described as partially-demineralized enamel typically experiencing mineral loss of 30 to 50%.[12, 14] Early caries lesions app ear to the naked eye as opaque white spots on the enamel surface. Microstructurally, the decay produce pitted surfaces. Caries lesions develop over a relatively long period of time, of ten years. The acid-buffering effects of saliva, removal of plaque from the tooth surface and regulating diet ary sugar intake limits the rate of lesion formation. Caries lesions serve as precursors to cavities when the decay progresses beyond the point of repair. Conditions encouraging the possible regeneration of decayed enamel are discussed in Section 2.4. Dental erosion is governed by similar solubi lity laws followed by caries lesion formation, as acid comes into direct contact with the enamel surface. Eroded lesions are formed by acids of nonbacterial origin, such as t hose typically found in food and drink. Lesions commonly appear on tooth surfaces that do not come in direct cont act with plaque. Early erosion, also known as enamel softening, extends to depths of 1 m.[16-19] Advanced-stage erosion is characterized by the complete layer-by-layer disso lution of enamel, which can make it difficult to measure in terms of relative mineral loss (Figure 2-7).[20] The tooth displays a polis hed, shiny appearance to the naked eye, making the condition difficult to detect (Figure 2-8). Eroded lesion formation occurs relatively quickly; it can occur w ithin a few minutes. Enamel (and dentin) demineralization can also be initiated by dental clinicians as a means to treat severe decay problems through an acid-etching process. In en amel, selective crystallite dissolution occurs, leaving behind a relatively porous surface. The toot h appears white and chalky to the naked eye,

PAGE 24

24 similar to a caries lesion white spot. Microstructurally, the enamel rod pattern is exposed, as seen by electron microscopy (Figure 2-9) Etched lesions typically extend to depths of 10 m. The increased surface area of etched dental tissue proves ideal for tooth bonding treatments, which are often initiated with p hosphoric or ethylenediamine tetraacetic acid (EDTA).[21] Treatment times may vary from a few seconds to several minutes, and is highly dependent upon the etchant used. While the chemical processes surrounding enamel dissolution are well-defined, there is a lack of information concerning the structural considerations involvedparticularly along the enamel surface. This may be due to the limited m eans available to access structural and chemical changes in tooth enamel at its most basic stru ctural levelalong the nanoscale. Current methods for the detection of tooth decay must be cons idered, as these are the primary ones used to monitor changes in enamel as a result of dissolution. Microradiography (MR) The use of x-ray absorption to examine the mineral structure of dental enamel was introduced by J. Thewlis[22] in 1940. This concept has evolved into the MR methods used today, long regarded as the standard means to assess diss olution of hard dental tissue. The most popular form of MR for decay detection is tran sverse microradiography (TMR). A CuK x-ray beam is directed toward a thinly-sliced sample (<100 m for enamel) and a nearby calibration wedge. This set-up is enclosed in a light-proof casing, with radiographic film placed beneath the sample and wedge.[23] Monochromatic x-rays interact with the sample, wedge and film to produce a microradiographic image. Microdensitometry traci ng of the image allows calculations of the gray levels present in comparison to a standard. Lighter areas on the micrograph indicate intact enamel; darker areas correspond to those of defi cient mineral content. Data indicating enamel mineral loss ( Z) and decay lesion depth (Ld) can be extracted. The mine ral volume %, V(x), at

PAGE 25

25 a specified lateral position is calculated using a formula developed by Angmar et al.[24] (Equation 2-2), where An,slice(x) is the x-ray absorbance of the tooth at a specified lateral V(x) = 100 (An,slice(x) / t) o ) / (m o) (vol %) (2-2) position, t is the thickness of the tooth sample, m is the linear x-ray attenuation coefficient of HAP, o is the attenuation coefficient of the organic component of the tooth. A typical MR plot modeling a subsurface caries le sion is shown in Figure 2-10. This concept has been expounded upon and extended to clinical use in the form of de ntal x-rays. This techni que serves as a great aid to visual and manual evalua tions of dissolved enamel. As caries lesions initiate and progress beneath the tooth surfac e, this method has proven to be an effective means of identifying this fo rm of tooth decay. A major limitation of the TMR technique is detection resolution. TMR is unable to observe enamel mineral changes less than 10 m from the tooth surface due to densitomete r slit restrictions and sample curvature.[25] This poses a problem when attempting to evaluate the earliest stages of caries development and erosion. At best, MR can indirectly measure structural changes in demineralized enamel on the micron scale. Profilometry Profilometry, or surfometry, is a method t ypically used to measure erosion along the enamel surface.[23, 26-28] Traditional experiments have used a diamond stylus, tens of microns in diameter, to scan across the sample surface at a specified rate. A comparison of profilometric traces produced by both untreated and eroded sample surface s cans can provide a lesion depth profile with micron resolution. Profilometric expe riments conducted with a laser stylus provide the added benefit of a potential three-dimensional interpretati on of the erosion lesion due to focusing and maneuverability of the laser beam, an example of which is shown in Figure 2-11. The laser beam does not physically touch the sample surface, eliminating potential damage

PAGE 26

26 through abrasion. Conditions for the profilometric study of demineralized enamel are a flat, baseline surface with reflective properties. In examining tooth enamel, the latter requires a shiny, metallic coating. Although profilometry has proven to be a quick, reliable method to measure mineral surface topographical changes, it also on ly gives an indirect measurement of nearsurface structural changes. Mechanical Testing Microhardness testing has long been an acceptabl e means of assessing mechanical property changes upon enamel demineralization.[29] Vickers or Knoop diamond indenters are applied under specified load and time conditions to a tooth sample. The size of the indent is measured. In surface microhardness testing, the indenter is app lied perpendicular to th e flat, polished, outer surface of the tooth. The indent size alone can gi ve a relative indication of the extent of demineralization, but must be correlated with a quantitative techni que (typically MR) to accurately assess mineral loss.[25] In cross-sectional microhardness, a Knoop indenter is typically applied to a sample cut parallel to the tooths biting surface. A mineral profile of enamel from the outer surface of the sample inward can be obtained. This provides an indirect method to quantitatively measure changes in mineral c ontent. Although providing more accurate mineral content data compared with surface microhardness, a major drawback is that the outer 25 m of the sample cannot be measured.[25] Mechanical changes in surface enamel due to early demineralization cannot be accurately determined. Although microhardness testing has been consid ered the standard in determining the mechanical properties of teeth, such a relativ ely large-scale method cannot resolve surface characteristics instrumental in early enamel dissolution. A more precise measurement technique is desirable. Nanoindentation is t ypically used to characterize the mechanical properties of thin films on the submicron scale. Under normal testi ng conditions, the material is subjected to a

PAGE 27

27 predetermined, submicron force applied at a contro lled rate. Indentation de pths of less than 1 m can be reached.[30] Real-time applied force and displacemen t data are recorded during penetration of the material, generating load-displacement cu rves. Data extracted from the unloading portion of the curves are used to calcu late properties such as hardness and reduced elastic modulus (Figure 2-12). Recently, nanoindentation has been gaining accep tance as a viable means of calculating the mechanical properties of dental tis sue. This technique is reported to produce indents in healthy enamel to depths of approximately 200 nm.[18] Sound enamel hardness and elastic modulus values from the literature based on the nanoindentation tech nique are listed in Table 2-2. Numerous nanoindentation studies have been attempted on demineralized enamel.[18, 31-33] These studies suggest that nanoindent ation data are superior to da ta obtained from the force, displacement and spatial resolutions of more traditional mechanical testing procedures when characterizing surface enamel. X-ray Diffraction (XRD) Improvements in decay detection resolution have been realized using XRD. Unit cell dimensions of the enamel crystalli tes, as well as an estimation of their physical sizes, can be calculated. During XRD analysis, a monochromatic b eam irradiates the samp le of interest. The x-ray beam penetrates the sa mple 0.1 mm, depending upon the material analyzed and the radiation wavelength.[34] Scattered x-ray photons from the sa mple are detected and converted into voltage pulses, producing a char acteristic spectrum of diffraction peaks.[35] Constructive interference of the scattered x-ray waves emitted at specific diffraction angles (2 ) and reflected from corresponding families of lattice planes must satisfy Braggs law to produce peak data (Figure 2-13). While the angular positioning of the spectrum peaks describe material crystal structure and phase symmetry, the peak geometries provide information about the crystallite

PAGE 28

28 sizes, lattice defects and interp lanar strain. The peak intensity provides information about the quantity of a particular phase in the material while peak width (indicated by the full width at half maximum, FWHM) determines relative crysta llite sizes and the presence of defects.[36] The intensity of a peak, indicating the amount of scattered radiation detected at 2 defines its height. In materials where interplanar strain is assumed negligible, the Scherrer equation may be used to determine average crystallite size (D) as follows D = / cos (2-3) where is the Scherrer constant (0.9 for HAP), is the x-ray wavelength, is the conditioned FWHM breadth of the diffraction peak and is the diffraction angle for a specified diffraction peak. Broadening of a diffraction peak suggests a decrease in average cr ystallite size and/or increase in lattice defects.[36] Factors such as the multiplicity eff ect of material planes of the same family encountering the incident x-ray beam at the same diffraction angle, th e structural effect of the material atomic positions on detected peak discernment and the polarization effect of x-ray wave propagation in relation to diffractometer detector position must be considered.[35] Conventional -2 XRD studies have been reported in the literature in the evaluation of untreated and demineralized enamel. Gawda et al.[37] positively correlated the velocity derived from ultrasonic measurements to crystallite size in dental tissue using XRD. The average crystallite size in enamel in the c-axis dire ction was found to be 45.4 5.2 nm and 26.0 4.4 nm in the a-plane. Crystallinity data were calcu lated from the (002) and (310) reflections, respectively. Dentin crystal lite size was reported as 23.9 13.0 nm along the c-axis and 6.6 1.2 nm within the a-plane. It should be noted that sample slabs were cut such that they contained overlapping enamel, dentin and cementu m portions of the tooth, thereby affecting the true calculated crystallinity of each designated ti ssue component. In an attempt to characterize

PAGE 29

29 surface vs. bulk enamel crystallites, Sakae[38] used XRD to analyze powdered enamel samples from both inner and outer enamel layers. Peak data from the (002) reflection of outer layer enamel showed an average FWHM value of 0.23, corresponding to an average unit cell length in the c-direction of 0.6884 nm. The (300) reflection, an indication of crystallinity in the a-axis direction, showed average FWHM and un it cell length values of 0.3000 and 0.9450 nm, respectively. Inner layer enamel exhibited slightly larger FW HM and unit cell length values compared with outer enamel, suggesting a decrease in enamel crystallite si ze with depth into the tooth in all directions. While conventional XR D techniques can provide a direct structural evaluation of tooth enamel, it lacks the surface se nsitivity to detect changes in the near-surface layers. Transmission Electron Microscopy (TEM) TEM characterization is based on examining the microstructure of enam el crystallites, as well as analyzing the lattice periodicity of thin se ctions of intact enamel The hexagonal shape of enamel crystallites has been verified through br ight-field imaging studi es (Figure 2-14). Of particular interest is the presence of a dark li ne bisecting the crystallite thickness in untreated enamel. White spots were also spo tted in the vicinity of the line[39]. Marshall and Lawless[40] observed a similar phenomenon, sugg esting that the line represents a planar defect along the (100) plane. Expounding upon this th eory is the suggestion of a localized modification in chemical composition, perhaps indicating the presen ce of a crystalline phas e. One suggestion is that this core region contains carbonate, an impurity known to make HAP more susceptible to chemical dissolution. There may be some validity to this theory in that the core regions of enamel crystallites are often the first to show signs of acid dissolution via TEM (Figure 2-15). Simmelink and Abrigo[41] reported centralized crys tallite dissolution proc esses with duration of

PAGE 30

30 acidic exposure until th e demineralization effects spread to the periphery of the crystal, leaving mere fragments of mineral behind. Structural inte grity is completely compromised at this point. Untreated sections of tooth enamel examined by TEM exhibited a lat tice structure similar to stoichiometric HAP in applicable studies. Brs and Hutchison[42] suggested the influence of enamel matrix proteins on cr ystallite growth. They found that growth control by the matrix proteins only occurs on the {120} planes. An examin ation of the matrix showed the presence of poorly-crystalline, HAP-like phases, which were shown to form grai n boundaries with the enamel crystallites. While the aforementioned techniques provide in sight into mineral changes that occur as a result of enamel dissolution, they are inadequate in examining such changes representative of that at the tooth-oral fluid in terface. Higher resolution surface analysis techniques would prove more helpful, and may assist in treatment of t ooth decay. As with bone, albeit through different mechanisms, partially-decomposed enamel may be regenerated. This process has been termed remineralization, and may occur in carious, an d in some cases eroded and etched enamel. Enamel Remineralization After periods of demineralization, in vivo the tooth has a propensity to naturally repair itself. Partially-dissolved enamel has the ability to be regenerate d to a near-healthy state via the transformation of enamel crystals. Partial en amel dissolution that le aves behind a suitable mineral substrate act as nucleation sites for gr owth. When the bulk saliva pH rises above the critical pH for enamel dissolution, there is a shift in equilibrium whereby the saliva becomes supersaturated with Ca2+ and PO4 3-.[14] Demineralization wanes and remineralization dominates, driving the calcium and phosphate ions to diffuse into the decomposed enamel to form new or modified apatite. The dynamic, interchangeab le, and sometimes concurrent processes of demineralization and remineralization in vivo conduct a balancing act in reacting to disruptions

PAGE 31

31 in the oral environment (Figure 2-16). The proce ss of apatite remineraliza tion has been described by researchers in one of three ways[39] Regeneration of partia lly-dissolved crystals Emergence of newly-formed crystals Additional growth of existing crystals Fluoride is important to the remi neralization process as it renders the enamel more resistant to acid dissolution. Fluoride absorp tion into the enamel has also been seen to enhance the chemical reactions that allow fo r apatite redeposition. The presen ce of relatively small fluoride ions in the apatite lattice reduces the free ener gy of the enamel surface thereby inhibiting the absorption of harmful salivary impurities.[12] Increased strength and resistance to fracture and wear of the enamel is due to the formation of fluorapatite (FAP), which dominates the upper 100 m of surface enamel.[43] The semipermeable nature of enamel allows for the absorption of Ffrom saliva to form FAP, strengthening the en amel structure and protecting the tooth against subsequent acid attack (Equati on 2-4). The most effective incorporation of fluoride Ca10(PO4)6(OH)2 + 2FCa10(PO4)6F2 + 2OH(2-4) in the oral environment is through the use of t opical agents, such as fluorinated toothpastes, mouthrinses and waters.[44] These agents provid e a reservoir of Ca2+ and PO4 3and Favailable for diffusion into surface enamel. At relatively high F concentrations, a CaF2-like layer of material may form on the surface.[45-47] Fluoride efficacy in enamel remineralization has been studied extensively thro ugh pH cycling studies.[48-50] The enamel sample is subjected to a series of demineralizing and remineralizing solutions over a specified period of time. Each solution is saturated with respect to HAP and FAP to simulate the formation of caries lesions in vitro .[51] Fluoride may be added in specified concentrations to the treatment solutions, or applied topically during cycling, to evaluate its reported positive effects on enamel stabili ty. This action most

PAGE 32

32 closely mimics the changes in pH experienced in vivo due to eating and drinking throughout the day. The phenomenon of remineralization, as well as the influence of fluoride, has been extensively reported in the literature. Amaechi et al.[52] reported an increase in mineral content after a1-hr exposure to orange ju ice in situ during subsequent e xposure to the participants own saliva. MR detected at least a 48% decrease in lesi on depth due to remineralization. Silverstone and Wefel[53] reported an increase in enamel mineral cont ent, driven by an increase in crystallite diameter, when subjected to calcifying remineralization solutions. Collys et al.[54] reported an increase in enamel microhardness as a result of remineralization via calcifying solution in the presence of low levels of fluoride or enha nced levels of phosphate. Conversely, Lippert et al.[33] reported no significant increase in softened enam el protection from toothbrush abrasion as a result of artificial saliva remineralization anal yzed via nanoindentation studies, suggesting the formation of an unstable remineralized enamel phase. The same group of researchers further substantiated their claim by conducting pH cycli ng studies and subsequent nanoindentation tests on the treated enamel. A collapse of the enamel surface layer due to subsurface demineralization was reported during testing, yielding results largely unrepresentativ e of the true nature of the remineralized layer.[55] Aoba et al.[56] observed a decrease in FWHM and an increase in the integrated intensities of the (004) and (310) re flections along the surface of artificial HAP pellets upon acid dissolution. A well-mineralized surface la yer formed on the pellets during subsurface dissolution, an occurrence that some researchers label as remineralization of the HAP surface layers. These results suggest an improvement in HAP crystallinity along the surface, driven by an overall increase in crystallite size. A study performed by Vieria et al.[57] used finely-ground human enamel samples subjected to fluoridated water from various regions for -2 XRD

PAGE 33

33 characterization. Results showed that peak br oadening along the (002) and (310) reflections occurred with fluoridation, suggesting a corr elation between crystallinity and enamel Fcontent. Many of the same characterization techniques us ed to evaluate demineralized enamel have carried over to the study of remineralized enamel. The process of biomineral remineralization is the least understood. While the beneficial effects of enamel remineralization has been obser ved through various experimental studies, there is debate within the scientific community su rrounding the validity of remineralization as explained by the structural regenerationor regrowthof demineralized enamel. Early TEM studies by Muhlemann et al.[58] concluded that a fine-grained, amorphous precipitate infiltrated intercrystalline spaces in enamel eliciting the possible reharden ing effects of remineralization. The amorphous precipitate was not identified in the study. The presence of this intercrystalline amorphous precipitate contradicts later claims of enamel mine ral phase transformation during remineralization. A study by Kawasaki et al.[59] suggested through high -resolution TEM studies that enamel transforms to a tetracalcium diphosphate monoxide phase upon remineralization. This conclusion was reached as a result of visual comparisons of TEM lattice fringe data with JCPDS data for HAP. At best, the literature cite s indirect means to deduce crystal growth as a result of remineralization based on microhardness, MR and XRD results. A true remineralization phase, aside from FAP, has yet to be identif ied. Additionally, there has been no definitive conclusion as to the role of fluorine in the remi neralization process. The question remains as to whether the enhanced properties of fluoridated enamel alone are capable of combating enamel dissolution, or if fluorine acts as a catalyst working in conjunc tion with specific conditions within the mouth to make the tooth more resistant to decay. Insight into the answers to questions such as these may be gained by examining what happens to enamel that has been remineralized

PAGE 34

34 within the surface layers. This becomes important when choosing ways to treat early signs of tooth decay in hopes of preventing cavitation.

PAGE 35

35 Figure 21. Schematic of an enamel crystallite. Figure 22. Schematic of enamel rods, hi ghlighting crystallite and rod orientation.[32]

PAGE 36

36 Figure 23. Photograph of a lower first prem olar human tooth slice showing enamel rod orientation.[60] Figure 24. Schematic highlighti ng major components of the tooth.[61]

PAGE 37

37 Table 21. Minor inorganic constitu ents in sound human tooth enamel. Constituent Concentration Range (wt%)[62] Substitutes For[63] Effect on Enamel Crystallinity[63] CO3 22.4.2 PO4 3Decrease F-* 0.05.9 OH-, ClIncrease Mg2+ 0.04.68 Ca2+ Decrease Na+ 0.17.16 Ca2+ None Cl0.16.7 OHNone *Present in highly variable amounts; F often considered a trace element. Figure 25. The idealized crystal struct ure of HAP, as viewed along the c-axis.[61]

PAGE 38

38 Figure 26. Factors contributing to caries lesion development in vivo.[64] Figure 27. SEM image of enamel eroded with citric acid for 30 min.[20]

PAGE 39

39 Figure 28. Photograph of facial erosion with shiny and smooth appearance.[4] Figure 29. Scanning electron microscope of enamel etched with EDTA. Arrow head indicates an enamel rod. Scale bar = 5 m[65]

PAGE 40

40 Figure 210. Schematic of an average enamel mineral content MR prof ile showing the mineral loss lesion ( Z) and the lesion depth (Ld).[66] Figure 211. Profilometric trace of eroded enam el lesion by a demineralizing agent for 30 min. Scale is in m[67]

PAGE 41

41 Figure 212. Typical load-displacem ent curve produced by nanoindentation.[68] Table 22. Sampling of sound tooth en amel hardness and modulus values. Lead Author Hardness (GPa) Elastic Modulus (GPa) Bhushan, B. (Ed.) [69] 2.7 5.2 89 106 Finke, M [18] 3.51 0.90 102.45 18.39 Habelitz, S [32] 3.9 0.3 87.5 2.2 Barbour, M E [31] 4.74 0.14 104.8 2.8 Lippert, F [33] 4.55 0.17 106.0 2.8

PAGE 42

42 Figure 213. XRD schematic illustrating Braggs Law.[70] Figure 214. TEM image of a cross-section of an enamel crystallite c-axis. Arrow points to a central dark line.[39]

PAGE 43

43 Figure 215. TEM image of cross-sections of demineralized enamel crystallite caxes.[71] Figure 216. Schematic illustrating the balanc e of demineralization and remineralization processes with regard to tooth decay.[64]

PAGE 44

44 CHAPTER 3 EXPERIMENTAL METHODS Sample preparation of the dental enamel used in this experimental study will be discussed in this chapter. The rationale behind the sel ection of demineralization and remineralization agents is explained. Finally, the characterizatio n techniques used to identify structural, morphological, mechanical and chemical chan ges in treated enamel are highlighted. Enamel Sample Preparation Extracted human incisor teeth were collected fr om the University of Florida dental clinics for use in this research. The dental histor y of the tooth donors was unknown. The teeth were sterilized by gamma radiation, using a Co60 radiation source at 0.5 Mr ad for at least 9 hours.[72] After sterilization, the te eth were stored in 0.1% thymol until needed. The sterilized teeth were se t in a cold-cure epoxy (Epofix, Cleveland, OH) for ease of handling during sample preparation. First, the extr aneous root sections of the teeth were ground away using 60 grit SiC abrasive paper. The teet h were positioned labial-side down to allow for easier access to the enamel surf ace. The epoxied samples were then successively ground with SiC abrasive papers to a 1200 grit fini sh. Alumina slurries, ending with a 0.05 m grit suspension, provided a highly-polished, flat enamel surface of approximately 5.0 mm2. Care was taken to remove at least 100 m of enamel, e liminating any traces of FAP for standardization purposes (Appendix A).[73] Finally, the samples were sliced into 2.0 1.0 mm thick slabs with a low-speed diamond saw. These highly-polished enamel samples were termed as-received, as they had not yet been subjected to any chemical surface treatments (Figure 3-1). The teeth were then ultrasonically cleaned with distilled wate r and stored in a hydrated environment, awaiting characterization or further treatment.

PAGE 45

45 Enamel Surface Treatments The enamel slabs were subjected to deminera lization and remineraliz ation agents, chosen based on their noted contributory effects to mine ral changes in the oral environment. Diffusion of Ca2+, PO4 3-, Fand other impurity ions into and out of tooth enamel is primarily governed by the pH solubility of enamel. Th e critical pH for enamel disso lution, which is approximately 5.5, drives ion diffusion such that it defines a dividing line between superand undersaturation of the oral fluids with respect to the enamel. Degrees of oral fluid saturation with respect to HAP and FAP as a function of pH at salivary concentratio ns of Ca are indicated in Figure 3-2. When the pH of the fluid falls well below the critical pH value, typically in a range of 1, undersaturation with respect to both HAP and FAP is observed.[74] In this pH range, the protective action of fluoride is not strong enough to prevent enamel di ssolution. Dental erosion, or etching, occurs. As the fluid pH increases to a range of approximately 4.5.0, it is un dersaturated with respect to HAP but supersaturated with respect to FAP. Ca ries lesions, defined as subsurface zones of partially-demineralized enamel, can form in this region. A relatively dissolution-resistant surface layer comprised of FAP forms over the lesion. Th e formation of this su rface layer is often thought to represent a form of remineralization, where apatite crystal growth is reportedly enhanced by the presence of F. These basic di ssolution principles govern the action of enamel lesion formation and have been considered in ch oosing treatment solutions for this research. Demineralization 0.12M Ethylenediamine tetraacetic acid (EDTA) In addition to pH considerations, the met hod and extent of enamel dissolution in acid solution is a function of acid type, relevant ion concentration (particularly that for Ca2+, PO4 3and F-) and duration/frequency of attac k. The chelating ability of certain acids also has an effect on apatite mineral stability.[75] Chelation, the calcium-binding ab ility of a compound, is usually

PAGE 46

46 prevalent in organic acids containing more than one car boxyl (-COOH) group. In the oral environment, chelating acids bind to the calc ium ions present in saliva to encourage undersaturation and subsequent enam el dissolution. The effect of pH for etching is negligible in this case. The greater the chelating ability of the acid, the more extensive the enamel demineralization. The action of chelators mimic th e effects of typical acid etching on dental tissue.[76] Ethylenediamine tetraacetic acid (EDTA) ha s been extensively used as a conditioner for enamel and dentin for adhesive bonding purposes.[65, 77-79] In this study, 0.12M EDTA (pH = 7.025 0.035) was used to produce etched enamel lesions.[79, 80] In a typical experiment, as-received enamel slabs were immersed in at least 25 ml of EDTA solution at time interv als up to 60 min. The solution remained at room temperature during the treatment, without agitation. Longer etch times were used to examine the demineralization microstructural pattern and progre ssive dissolution of enam el as a result of the EDTA treatment. Buffered lactic acid Lactic acid is a by-product of oral bacterial interaction with fermentable carbohydrates during caries lesion formation in vivo Demineralization recipes co ntaining lactic or other organic acids and various concentrations of cal cium, phosphate and low levels (ppm) of fluoride have been developed to produce solutions en couraging the formation of caries lesions.[73, 81-84] Buffering of the lactic acid solution was inte nded to mimic the buffering effects of saliva in vitro In this study, a buffered lactic acid demine ralizing solution based on a formulation by ten Cate and Duijsters[51] (pH = 4.55 0.014; [F] = 0.5 ppm) was used to simulate bacteria-induced demineralization in vitro (Table 3-1). As-received enamel slabs were immersed in at least 25 ml of solution for 30 min at human body (37C) and room (20C) temperatures, without agitation. A

PAGE 47

47 treatment time of 30 min was selected as it has been cited as being the amount of time needed for acid clearance in the mouth after a meal.[49] Demineralization effects we re intended to represent those seen along the enamel surface as a re sult of early caries lesion development. Remineralization Calcifying solution Saliva is known to aid in the remineralizati on of enamel through its buffering effect on acids found in the oral environment and when s upersaturated with calcium and phosphate with respect to demineralized enamel. It al so contains trace amounts of fluoride ( 0.1 ppm), which is widely accepted as a means to shift the decay pro cess to favor remineraliza tion and to facilitate the diffusion of calcium and phosphate ions fr om the saliva back into the enamel crystal structure.[52, 85, 86] This action may also enc ourage the production of FAP.[87, 88] The use of artificial saliva or calcifying solutions in vitro has introduced the saliva ef fect into the evaluation of eroded and caries-induced demineralization. These solutions often contain an organic buffer, such as N-2-hydroxyethylpiperazine-N-2-ethanes ulfonic acid (HEPES), and supersaturated concentrations of calcium and phosphate. In this study, a calcifyin g remineralization solution[51] (pH = 6.97; [F] = 0.1 ppm) was used to simulate artificial saliva effects in vitro (Table 3-2). As-received enamel slabs were immersed in at least 25 ml of solution for up to 8 h, without agitation. The long treatment time was chosen to simulate the remineralization effects of the saliva overnight. MI Paste topical cream Topical fluoride treatmentsnamely dentifrices, mouthrinses and gelshave proven effective in protecting the tooth against enamel dissolution. Containing relatively high concentrations of fluoride (up to tens of thousands of ppm), some treatments have the ability to form a calcium fluoride-like substance in plaque and on surface enamel to provide a reservoir of

PAGE 48

48 enamel-rich ions available for remineralization.[44-46, 89] MI Paste (CG Corporation, Tokyo, Japan), a topical cream cont aining a casein phosphopeptide-am orphous calcium phosphate (CPP-ACP) complex, provides a rese rvoir of freely-available cal cium and phosphate ions when adhered to the surface of the tooth. CPP has the ability to bind to amorphous calcium phosphate, stabilizing these ions in solution.[90] The bioactive ACP works to release these ions from enamel to maintain a supersaturated state near the enamel surface.[91] Based on Recaldent (originators of CPP-ACP complex) technology, MI Paste may serve as a bi oavailable source of calcium and phosphate without the influence of fluoride. Va rious studies have reported its efficacy in the remineralization of eroded and carious subsurface lesions.[91-93] MI Paste was applied to demineralized enamel slabs for up to 10 min, in 5 min intervals. Following each treatment, the cream was removed, rinsed and sonicated fo r 1 min in distilled water. Enamel Characterization Acids involved in enamel dissolution must firs t interact with and penetrate the surface of the tooth in vivo A mineral density gradient forms as a re sult, with a decrease in mineral content in areas affected by the acid as it travels more deeply into the dentition. The structural integrity of the demineralized enamel layers is compromised with respect to bulk enamel, causing a discrepancy in the material properties displayed by both regions. Early enamel dissolution, which produces softened enamel, is defined as demi neralization that extends 2 to 5 m from the physical tooth surface.[17] Remineralization of early deminera lized lesions works to regenerate lost mineral, whose effects can be fully re alized within the softened enamel regime. Dimensionally, treated enamel is on the scale of th at of a thin film, ranging from an atomic layer to several micron thicknesses. Thin films have found application in the semiconductor, microelectronics, communications and material coating industries where the use of small-scale

PAGE 49

49 components is common.[94-97] As such, the samples used in this research may be analogous to the structure of a polycrystalline thin film/substrate co mbination in that the treated surface layer is representative of a thin film atop an untreated, bu lk enamel substrate (Figure 3-3). To accurately evaluate treated enamel, characterization methods sensitive to surface inter actions are desirable. Various high-resolution characte rization techniques were used to describe changes in surface morphology, structure, mechanical properti es and chemistry after the demineralization and remineralization of tooth enamel. Grazing incidence x-ray diffraction (GIXD) and x-ray reflectivity (XRR), techniques widely used in thin film analyses, were used to assess structural changes in the surface layers of enamel upon tr eatment. High-resolution x-ray scattering results were corroborated with nanoindentation studie s and confocal Raman microscopy performed on select samples to assess changes in mechanical properties and su rface layer chemistry, respectively. Surface imaging was done using sc anning electron microscopy (SEM) and atomic force microscopy (AFM), performed to examine the topographical features of as-received and treated enamel samples. High Resolution X-ray Scattering Advanced x-ray scattering methods were used to assess crystallinity changes in tooth enamel as a result of demineralization and remi neralization treatments. X-ray diffraction studies were performed using an XPert Pro Materi als Research Diffractometer (PANalytical, Westborough, MA). With regard to this resear ch, its high resolution capabilities were most suitable for XRR and GIXD phase analyses. Us e of the XPert provides a nondestructive evaluation of the structural, and to a degree, chemical characteristics of the material of interest. The XPert Pro uses a four-axis goniometer system that allows for precise sample, beam and detector placement for material analysis. The 2 parameter, which defines the angle between the incident and diffracted beams, controls the positioning of the detector. The omega ( )

PAGE 50

50 parameter defines the angle between the sample and incident beam. The rotation parameter, phi ( ), turns the longitudinal plane of the sample. The tilt parameter, psi ( ), allows for horizontal tilting of the sample. The latter two parameters are controlled by the movement of the sample stage, which has a 0 range of motion (Figure 3-4). The x-ray source is a tungsten filament, which interacts with a copper anode for x-ray production. Primary and secondary optics are placed in th e line of the beam for conditioning purposes. The primary (incident) beam optics prepares the x-ray beam for sample interaction as it leaves the filament tube. The x-ray mirror device suppresses the K radiation to produce a beam consisting primarily of K radiation. Reflection of the beam onto a parabolic mirror converts it into a quasi-parallel state. An automatic attenuat or is attached to the x-ray mirror device to prohibit damage to the x-ray detectors; it can be set to allow no more than 500,000 counts per second (cps) at its maximum. Afte r interactions with the sample, the beam is further conditioned through the secondary optics. A pa rallel plate collimator, used to further decrease x-ray beam intensity, attached to a xenon detector completes this set-up. A beam mask may be inserted into the secondary optics set-up to decrease the height of the beam, if necessary. X-ray reflectivity (XRR) Specular XRR relies on the constructive interference of x-rays reflected at layered interfaces within a sample, distinguishable by varying material electron densities. While remains below a critical angle c, specular reflection of the x-ray beam from the sample occurs. As increases to values above c, x-ray penetration occurs. The critical angle for total x-ray reflection is sample-specific and has a value of 0.05.50 fo r most materials.[34] This value is dependent upon the sample electron density and x-ray wavelength. Determination of c is based on the refractive properties of the sample. The refractive index (n) is defined as n = 1 (3-1)

PAGE 51

51 where is a dispersion term based on the electron density of the sample defined as = ( 2/2 ) re NA (Z+/A) (3-2) where is the x-ray wavelength, re is the classical electron radius, NA is Avogadros number, is the mass density, Z is the atomic number, is a dispersion correction factor and A is the atomic mass. Using Snells law of refraction and assuming nair ~ 1.0, nsample < 1.0 and total reflection of the x-ray beam ( refraction = 0), c is defined as c = 2 (3-3) c has been approximated at 0.25 for HAP.[98] Below c, the x-ray beam penetration depth (d) is calculated by d = / 2 ( c 2 2)1/2 (3-4) Above c, d is calculated as d = 2 / m (3-5) When the incidence and reflected angles are the same, differences in electron densities among the sample surface and internal layers can be detected.[99] The resultant XRR curve can be matched against a simulated mathematical model to estimate surface and/or multilayer thickness, roughness and mass density. The periodicity of interfer ence fringes often found along the XRR curve determines the thickness of a single layer (tl) based on the relationship tl = / 2 (3-6) Surface roughness enhances diffusive x-ray scatteri ng, which lowers the overall intensity of the specular XRR curve. The roughness between layers and a decrease in mass density are indicated by a decrease in the amplitude of the interference fringes. The calculated density of each sample

PAGE 52

52 layer can be exacted from the and c values calculated from Equations (3-2) and (3-3), respectively.[100] Grazing incidence x-ray diffraction (GIXD) During GIXD, the x-ray beam penetrates the sa mple a few thousand angstroms, irradiating a large area along its surface. In contrast to conventional -2 and similar to XRR studies, the beam is directed toward the sample at a small incidence angle ( ) around c. A portion of the beam is diffracted (not reflected) by the sample and is used to genera te GIXD patterns of the surface layers of the material. With a GIXD measurement, remains constant as the x-ray detector moves in the plane of 2 Varying for each measurement creates a depth prof ile of crystallinity (peak) information within the surface region (Figure 3-5). A relatively large value (above ~7.0) above the critical angle will cause x-ray penetration into the bulk, in which case the bulk diffraction peaks may obstruct those of smaller intensity from the surface. The GIXD geometry does not allow the presence of strong diffraction lin es from the sample substrate (i.e., the bulk) from interfering with the spectra of the sample itself. Figure 3-6 displays its strength in resolving surface layer crystallinity in titani a deposited onto a glass substrate. The amorphous nature of the glass detected by conventional XRD methods produced a halo that overshadowed the crystallinity data exploited by GIXD in the same region. GIXD is cap able of detecting crys tallinity changes at depths of 10 nm from the sample surface. Experimental set-up In this research, copper K radiation ( = 1.542 ) was used. All measurements were made as continuous scans in the powder mode. The machine operated at 45 kV and 40 mA under the line focus of the x-ray beam. The Cu K attenuation length for HAP, which describes the depths from which diffraction data are ex tracted, is shown in Figure 3. For -2 measurements, a 1/2

PAGE 53

53 divergence slit was used. A step size of 0.02 and time per step of 7.0 s were used to produce a high signal to noise ratio. For GIXD, the same scan parameters were used, except for the use of a 1/8 divergence slit. In addition, a succession of values between 0.15 and 0.4 were used to monitor changes near the surface of the samples. Grazing incidence beam values of 0.8 and 1.0 were also run and extrapolated to -2 to trace crystallinity changes into bulk enamel. Portions of the resultant spectra were modeled w ith ProFit curve fitting software.[101] For XRR measurements, a 1/32 divergence and 0.1 mm r eceiving slits were use d. A step size of 0.005 and time per step of 6.0 s within the s can range of interest were used. Nanoindentation Nanomechanical testing on toot h enamel can be used to quantify changes in surface structure. A diamond indenter is typically used for probing. The probe contacts the sample surface with a predetermined force at a particular rate. A force-displacement curve is generated, which can be used to determine the mechanical properties of the sample.[69] Using a power law relationship, the applied load (P) is determined as P = A(h0-hf)m (3-7) where A is the surface area, h0 as the initial penetration depth, hf as the final depth, and m = 2 (conical). The stiffness (S) of the material is defi ned as the derivative of the load with respect to the penetration depth at the maximum load, Pmax S = dP / dh at Pmax (3-8) The contact depth (hc) of the probe at a particul ar instant is defined as hc = hmax 0.75(Pmax / S) (3-9) Finally, the hardness (H) and reduced modulus (Er) values were found by H = Pmax / Ahc (3-10) Er = [S ] [2 (Ahc)] (3-11)

PAGE 54

54 The reduced modulus, which accounts for the mechanical properties of both the sample and the indenter, can be relate d to Youngs modulus by 1 / Er = (1 s 2) / Es + (1 i 2) / Ei (3-12) where s and Es are Poissons ratio and Youngs modulus of the sample and i and Ei are those of the indenter, respectively. Poisson s ratio for HAP is estimated at 0.28.[102] For a standard diamond indenter, Ei is 1141 GPa and i is 0.07. A TriboIndenter (Hysitron, Mi nneapolis, MN) was used for nanomechanical testing of as-received and treated enamel slabs. A Berkovich diamond indenter was used as a probe for sample testing. Nanoindentation curves illustrati ng changes in hardness and elastic modulus as a function of contact depth were used to quantif y mechanical changes al ong the enamel surface. Nanomechanical calculations were executed ba sed on methods refined by Oliver and Pharr.[30] Confocal Raman Microscopy A WITec alpha300R confocal Raman microscope (Ulm, Germany) was used to derive chemical information in the form of Raman spect ra and chemical maps from surface and depth profile scans of as-received and treated enamel. This nondestructive technique focused a monochromatic laser light onto a sample to indu ce inelastic scattering of its photons, effectively measuring the vibrational energies of the molecules within the sample.[103] Averaged spectra of Raman bands, corresponding to vibrational freque ncies unique to spec ific functional groups within the enamel surface layers were generated (Table 3). Th e spatial distribution of areas representative mineral and organic phases in the sample, from which Raman spectra can be derived, were mapped with the confocal portion of the microscope. Scanning Electron Microscopy (SEM) SEM was the preferred method used for the imaging enamel surface features.[65, 104-106] A JEOL SEM 6400 (Tokyo, Japan) was used to exam ine micron-scale variations in enamel upon

PAGE 55

55 treatment. Sample preparation consisted of dehydrating and coating all samples with Au-Pd alloy. Secondary and backscattered electron images were recorded with an accelerating voltage of 5 kV. Images were taken in areas high lighting the effects of demineralization and remineralization on the enamel rods, the promin ent structural compone nts of tooth enamel. Additional changes in enamel microstructure were imaged with a JEOL JSM-6335F field-emission SEM (FE-SEM). Its high-density electron beam at a relatively small spot size provided maximum brightness to discern features as small as 5 nm at 5 kV. Atomic Force Microscopy (AFM) For imaging as-received and select treated samples, a Dimension 3100 AFM (Veeco, Plainview, NY) was used. Since coating the sample is not necessary for AFM analyses, better contrast and reso lution of finely-polished enamel sa mples could be realized. For this research, the AFM was operated in the tapping mode. Average RMS roughness values were extracted in representative ar eas along select samples. All analyses were performed under atmospheric conditions using standard Si3Ni4 tips. Attempted TEM Characterization TEM analyses of as-received and chemically-treated enamel were attempted for this research. Due to difficulties encountered in sample preparation for this method, results were not obtained. The cross-sectional examination of inta ct enamel from the surface toward the bulk was desired. A depth profiling of changes in electr on density and/or lattice mismatch would have been correlated with high-resolution x-ray scatte ring data. Energy dispersi ve spectrometry (EDS) were to be attempted with the surface region of the enamel samples to more accurately pinpoint any phase changes that may have occurred during enamel treatment. Although the preparation of thin enamel samp les for TEM analysis have been prepared most often through ultramicrotomy, the high deform ation induced often leads to cracking of the

PAGE 56

56 brittle enamel matter. To help eliminate this problem, focused ion beam (FIB) milling has been a proposed method for sample thinning. FIB has b een attempted on synthetic apatite materials,[107] and recently on actual tooth enamel[108, 109] and dentin[110, 111] tissue. Although HAP is cited as being resistant to ion beam damage, le ss stable apatitic stages may amorphize.[112] This condition may alter the chemistry of remineralized enamel surfaces. As the FIB process is time-consuming and takes great precision and accu racy, we had difficulty developing a sample preparation method compatible with the Strata DB235 (FEI, Hillsboro, OR) system available for use. Due to lack of FIB sample availab ility, TEM analysis was not attempted. Statistical Considerations Due to the limited number of suitable teeth available for analysis, significant statistical analyses could not be achieved. For crystallite si ze calculations, at least three samples from each treatment stage were analyzed. Results were plotted in a line graph showing an averaged crystallite size with error bars indicating the standard deviations of th e sizes calculated. While the results may not be statistica lly relevant, trends in enamel crystallite behavior could be observed. Differences in between as-received and treated enamel groups analyzed by nanoindentation and XRD calculation of average crystallite size were determined with 95% confidence.

PAGE 57

57 Figure 31. Extracted human t ooth and epoxied enamel slab. Figure 32. Solubility isotherm s expressing the relationship between the enamel mineral ions (Ca) and pH at saturation for HAP and FAP. The arrow marks approximate [Ca2+] in oral fluids.[74]

PAGE 58

58 Table 3-1. Composition of buffered l actic acid demineralization solution.[51] Component Concentration (mM) CaCl2H20 2.2 NaH2PO4H20 2.2 85% C3H6O3 (lactic acid) 50 F (via NaF) 0.5 ppm pH ~ 4.5 Adjusted with NaOH Table 3-2. Composition of calcify ing remineralization solution.[51] Component Concentration (mM) CaCl2H20 2.2 KH2PO4 0.9 KCl 130 HEPES 20 F (via NaF) 0.1 ppm pH ~ 7.0 Adjusted with NaOH Figure 33. Schematic illustrating the analogy of an enamel surface layer on an untreated, sound enamel substrate to a thin film model.[17]

PAGE 59

59 Figure 34. PANalytical XPERT MRD PRO f our-axis set-up. A) Goniometer and secondary optics. B) Goniometer angle designations. A

PAGE 60

60 Figure 35. GIXD geometry.[113] Figure 36. Comparison of XRD spectra of TiO2 on a glass substrate produced by conventional -2 geometry and GIXD surface-specific geometry.[113]

PAGE 61

61 Figure 37. X-ray attenu ation length of Cu K radiation in HAP. Table 3-3. Raman frequencies and PO4 3vibrational modes associated with tooth enamel.[103] Vibrational Mode Frequency (cm-1) 3 1045 1 960 4 590 2 430

PAGE 62

62 CHAPTER 4 ENAMEL CHARACTERIZATION Introduction An adequate assessment of changes in surface-treated enamel requires comparisons to sound, untreated enamel. Characterization of such a sample provides a baseline for relative comparisons of structural, mechanical and chemical changes in the enamel upon demineralization and remineralization. The microstr ucture of enamel, with enamel crystallite dimensions as small as tens of nanometers, requires careful consider ation and precision in describing its material properties. The hierarchal microstructure of enamel results in anisotropic mechanical and structural properties due primarily to its preferred crystallite/enamel rod orientation. Enamel crystallites bundled to form enamel rodsare orthogonally-po sitioned relative to the outer tooth surface. A pliable, organic matrix cushions the rods a nd absorbs compressive forces induced by chewing and biting. In essence, enamel is analogous to a more generalized description of a fiberreinforced composite.[114, 115] The literature cites evaluations of mechanical properties parallel and orthogonal to the occlusal (biting) surface of human molars, schematically represented in Figure 4-1. Cuy et al.[116] found slight variations in hardness and elastic modulus along the tooth surface, with the highest values measured along the occlusal surface. Interestingly, variation from the enamel surface to DEJ varied greatl y. Mechanical property dependence on orientation provides an example of the influence of the crys tal structure of tooth en amel on tooth form and function, a topic explored throughout this research. In addition to local variance in mechanical pr operty values, changes in crystallite size and overall enamel thickness with position were also reported. Low[98] reported that enamel had a graded nature in terms of crystallite size, with an average decrease in size from the tooth surface

PAGE 63

63 inward to the DEJ. For human incisor teeth, a difference in enamel thickness was reported by Gillings and Buonocore,[8] with the greatest thickness at approximately 2 mm from the occlusal surface of the tooth. The natural variance seen in th e tooth, a true biological material, will affect the treatment and characterization results obt ained from each enamel sample. In this experimental study, as -received human incisor enamel slabs were used to determine baseline enamel material properties. Surf ace morphology was examined using AFM and SEM. Comparisons were made using the conventional -2 XRD and GIXD techniques to examine the crystal structure within the bulk and surface laye rs of enamel, respectively. The nanoindentation technique was used to provide nanohardness and el astic moduli data. Confocal Raman data were extracted to provide a depth profile and chemi cal data within the enamel surface layers. In combination, these surface characterization techni ques provided an adequate description of prominent characteristics of tooth enamel affected by demineralization and remineralization. Materials and Methods Enamel slabs were prepared and characterized as outlined in Chapter 3. AFM Two-dimensional images and representative roughness (RMS) values were taken of as-received enamel using the AFM. Observatio n of the surface morphology, without the possible obstruction of coatings required fo r SEM analysis, was exploited. SEM Secondary electron images were taken of as -received enamel using the FE-SEM. Image magnification was greater than the sizes of the primary enamel microstructure, namely the enamel rods.

PAGE 64

64 XRD Conventional -2 XRD scans were obtained for as-received enamel. A 2 scan range of 20 to 55 was selected to display a full ra nge of prominent enamel diffraction peaks. High resolution x-ray scattering was performed using the XPert system. For generation of the XRR curve, a 2 scan range of 0.1 to 2 was selected to highlight the critical angle for total x-ray reflection. For GIXD, a 2 scan range of 25 to 35 was selected to highlight the most prominent diffraction peaks reportedly affected by the demineralization and remineralization processes to be used in subsequent experiment s. Special emphasis was placed on the 25-27 2 range, the location of (002) reflection indicative of the enamel crystallite c-axis direction. Data analysis was performed with software used in conjunction with the XPert system. Conventional XRD and GIXD scans were evaluated for crystallinity using the ProFit program, a curve fitting application capable of extracting accurate FWHM data. Nanoindentation The Triboindenter system was used to pe rform nanomechanical te sting. A three-sided, pyramidal Berkovich indenter tip was used to probe the enamel slabs. With a total included angle of 142.3, half angle of 65.35 a nd radius of curvature betwee n 100 nm and 200 nm, this tip allows for maximum spatial resolution when probing surface and localized structures within highly-mineralized enamel. A tip area function calibration was performed on a fused quartz standard, which determined the projected tip contact area within the material at peak load. Indent grid patterns, 10 x 10 in size with indents spaced 10 m apart, were applied to the central region of each sample. A trapezoidal load function, with drift control, was used to minimize any creep effects produced in softer areas of the tooth. A typical experiment in cluded a constant load application of 1500 N based on the procedure outlined in Habelitz et al.[32]

PAGE 65

65 Confocal Raman Microscopy Confocal imaging provided X-Y surface and X-Z depth profile images of as-received enamel. The images were color-coded to highlig ht the location of specific phases within each sample. A 532 nm laser was used for bond excita tion, along with a 100x air objective lens (N.A. = 0.9) for focusing. An averaged Raman spectrum was extracted to highlight the vibrational energies of the phases present in tooth enamel. Results Figure 4-2 is an AFM image of as-received enamel. There are traces of debris and the presence of fine polishing lines. No discernable microstructure was observed. RMS roughness observed over the entire image was 9.66 nm. An FE-SEM image showed outlines of aggregated enamel crystallites in Figure 4-3. Cracks were visible due to dehydration of the enamel sample. Conventional -2 XRD spectra of as-received enamel and a powdered NIST HAP standard (#2912) were in agreement in terms of visi ble peak position according to JCPDS #09-472,[117] although many of the enamel peaks were not resolved (Figure 4-4). The full enamel XRD pattern was much lower in intensity than that for th e HAP standard, the pres ence of a halo among the low-order reflections. The -2 and GIXD spectra comparison in Figure 4-5 showed even fewer peaks in the latter spectrum, with the emergen ce of the (200) reflection. The critical angle for total x-ray reflection, as estimated from the XR R curve in Figure 4-6, is approximately 0.25. Each (002) reflection was modeled with the ProFit software to extract accurate FWHM values, as illustrated in Figure 4-7. A depth profile highlighting changes in FWHM along the c-axis direction from the enamel surface into the bulk is shown in Figure 4-8. Nanoindentation testing revealed as-received enamel hardness and elastic modulus values of 4.86 0.44 GPa and 95.87 5.58 GPa, respectively. These values were derived from averaged load vs. displacements plots, an example of which is seen in Figure 4-9.

PAGE 66

66 Confocal Raman studies of as-received enamel showed an X-Y surface scan consisting of an array of enamel rod outlines. They appeared to correspond to the terminal ends of the vertically-aligned enamel rods seen in cross-section in Figur e 4-10. The images were colorcoded in orange by averaging the highest-intensity 1(PO) band across the images. The black areas within the mineral regions of the images represent the organic matrix. The averaged Raman spectrum indicated the presence of rather shar p phosphate band vibratio nal peaks at 432/cm, 586/cm, 959/cm and 1070/cm (Figure 4-11). A C-H stretching band, which indicates the presence of organic material, was seen. Discussion Tooth Enamel Defined FE-SEM imaging showed the terminal ends of relatively intact enamel crystallites, with surface flattening via polishing working to expos e preferential orientation along their c-axis directions. AFM imaging did not resolve crystal lite features due, in part, to the scan size selected, a decision that will become more obvi ous in later experimental studies. Crystal orientation is further corroborated by the presence of only a few reflections of the total number possible in the -2 spectrum of the enamel. This indicate s texturing within the material. The (002) reflection has the highest intensity, indicating that the greatest number of x-rays are reflected from the (002) planes. Th is provides further evidence of the crystallite caxis direction as the preferred orientation. A lack in overall enamel spectrum intensity as compared with pure HAP may indicate the relatively poor crystallin e quality of enamel and x-ray scattering interference from the organic material present in enamel. The organic material is indicated by the presence of a broad amorphous peak, or halo, in the -2 spectrum. The benefit of the GIXD technique is illustrated in Figure 4-5, where the presence of the (200) reflection within surface enamel is resolved. This peak cannot be di scerned with conventional XRD techniques. A

PAGE 67

67 drawback is a further decrease in peak intensity of the GIXD spectrum. This is a result of the shallow x-ray attenuation length used in GIXD. Fewer crystallite s are available for scattering, therefore producing peaks of lower intensity.[36] The XRR curve will serve as a baseline for examining shifts in c after selected enamel surface treatments. This action is indirectly related to changes in enamel density as seen in Equation 3-3. Nanohardness and elastic modulus values obtain ed for both as-received enamel and dentin were comparable to values reported in the literature (Table 2-2). Acceptable data for evaluating mechanical properties at the nanoscale using this technique included devia tions from the average calculated values of no more than 10%. Fewer da ta were deemed acceptable in the analyses for dentin as opposed to enamel. This may be due to the relatively pliable surface observed for dentin.[118] Lateral variance in mechanical properties may be partly responsible for the error spread seen when applying nanoindentation stud ies to biological materi als in general. Confocal Raman microscopy further verified th e preferential orientat ion and structure of tooth enamel used for this research. Rama n laser scattering induced strong phosphate band excitation in enamel, a group of which are better suited for Raman enamel studies due to their high intensities relative to Ca-OH bands in the 300/cm-range.[119] The presence of sharp phosphate bands, especially at 959/cm, is an indicati on of relatively strong, intact enamel. Due to the anticipated high variability of enamel material properties and the precision with which the enamel samples were characterized, using the same tooth for each complete cycle of treatments and/or set of charac terizations is desirable. Characterization Concerns Biological materials introduce uni que challenges in experiment al execution. Tooth-to-tooth variability among individuals with in a sample population can be expected based on genetics and

PAGE 68

68 oral care over the lifetime of the teeth. Late ral and depth variability in enamel thickness, hardness and crystallinity for each enamel sample must be considered during data analysis. In addition to inherent structural variabilit y, experimental environment plays an important role. Whole teeth thrive in a warm (i.e., human body temperature of 37 C), moist environment in vivo Deviations from the tooths natural state were incorporated into this research. All experimentation was performed in vitro and at room temperature, unless otherwise specified. Due to high vacuum needs of the SEM, the enam el samples were dehydrated prior to coating. This made the enamel samples more susceptibl e to cracking. Enamel polishing was necessary to provide a relatively large, flat surface area wi th minimal roughness, important for instrument error reduction in high-resolution XRD and nanoindentation studies. Polishing also allowed for better standardization of the enamel studied in that the natura l curvature of the incisor teeth was flattened and outer layers of fluoridated enamel was removed. Unfortunately, polishing induces physical trauma to the tooth, introducing possibl e defects. Finally, tooth storage methods after polishing and prior to char acterization may affect enamel mate rial properties. Reports indicated that the storage of tooth enamel sections in deionized water causes a decrease of its nanomechanical properties by 50% after 1 week of exposure, as determined by nanoindentation studies.[120] Diffusion processes extracted the structurally-important Ca2+ and PO4 3ions from the enamel into the water, causing demineralizati on. As such, the enamel samples were treated and/or characterized shortly afte r polishing to decrease idle storag e time. When stored, the teeth were kept hydrated, although not fully immersed, in distilled water.

PAGE 69

69 Figure 41. Average values of enamel hardne ss (a) and elastic modulus (b) as reported earlier by other researchers for the mesial half of a maxillary M2 as determined by nanoindentation. The standard deviations fo r these averages range from 0.2 to 0.3 GPa for hardness and from 2 to 5 GPa for elastic modulus.[116]

PAGE 70

70 Figure 42. AFM images of as-receiv ed human tooth enamel. Scale = 20 m2 Figure 43. FE-SEM image of as-received enamel.

PAGE 71

71 0 600 1200 1800 2400 3000 2022.52527.53032.53537.54042.54547.55052.555Intensity (a.u.)Theta NIST Standard -HAP As-Received(002) (004) (213) (222) ( 202) (210) (310) (200) (300) (112) (211) Figure 44. Conventional -2 XRD comparison of as-received enamel and NIST calcium HAP standard #2910. Spectra vertically displaced for clarity. Pattern indexed according to JCPDS PDF #09-472. 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2022.52527.53032.53537.54042.54547.55052.555Intensity (a.u.)Theta T-2T omega = 0.4(200) (202) (213) (222) (211) (002) (112) Figure 45. Comparison of as-received enamel analyzed by conventional -2 XRD and GIXD ( = 0.4). Note emergence of (200) reflection.

PAGE 72

72 1.0E+01 1.0E+02 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.20.50.71.01.21.51.72.0log Intensity (a.u.)Theta Figure 46. Representative X RR curve of as-received enamel. 2-theta 25.4 25.6 25.8 26 26.2 26.4 Intensity 2100 2200 2300 2400 2500 2600 2700 -0.02 0.02 Figure 47. ProFit (002) peak modeling.

PAGE 73

73 15 20 25 30 35 40 0.150.20.250.30.40.81.0T-2TCrystallite Size (nm)Grazing Incidence Angle, ( ) Figure 48. Depth profile of th e average crystallite sizes along the c-axis from the enamel surface into the bulk. Figure 49. Representative force (load)-displacement nanoindentation curve for as-received enamel.

PAGE 74

74 Figure 410. Confocal microscopy images of as -received enamel. A) X-Y surface scan. B) X-Z cross-sectional scan. Figure 411. Averaged Raman spectrum of as-received enamel. Both apatitic and organic components are shown.

PAGE 75

75 CHAPTER 5 ENAMEL DEMINERALIZATION Introduction The study of enamel dissolution in vitro is performed by the intentional acid attack of the apatite surfaces. While conditions such as experi mental temperature, acid type and attack frequency/duration are regulated to mimic oral conditions as closel y as possible in caries lesions and fluoride efficacy testing, acid produc tion by bacterial metabolism of sugars in vivo can only be simulated. Other factors, such as the ever-changing concentrations of Ca, PO4, F and other trace elements in oral fluids and the buffering e ffects of the acquired pell icle absorbed onto the enamel surface, are difficult to model. A goal of this research is to examine structural changes in surface enamel upon demineralization at the submic ron level. As such, etched enamelwhich exploits the intentional acid attack of enamel for restorative purposesappeared a suitable means by which to examine enamel material property changes upon demineralization. For comparison purposes, buffered lactic acid was used as an additional demineralization agent to better simulate enamel dissolution activities in vivo. EDTA is an etchant used to condition toot h enamel and dentin for use in dental restorations. A highly-roughened, porous surf ace is produced, increasing the surface area available for adhesion of sealants and other bondi ng materials to enamel. In cases of extensive tissue damage, collagen tubule exposu re in dentin is needed to enha nce attachment to this softer region, creating a suitable interfacial layer.[77] 37% phosphoric acid at pH< 2.5 is most commonly used for etching purposes.[79, 121] A 30 s treatment is strong enough to penetrate the enamel and permeate into dentin, effectivel y causing collagen dissolution. Unfortunately, excessive acid penetration experienced at higher acid concentrations may create dissolution channels so deep to where the restorative mate rial cannot reach, introducing voids and potential

PAGE 76

76 leaking at the adhesive/ tooth bonding sites.[77] A milder acid treatment that encourages demineralization within the dentin tubules, leaving the remainder of the organic matrix intact is desired. EDTA can achieve this goal. This acid type is unique in that it can cause enamel dissolution at neutral pH through chelating. Some b acterial acids involved in the caries formation process were found to have chelating propertie s, although their degrees of influence on the enamel dissolution in vivo have not been well researched.[122] Buffered lactic acid solutions have been exte nsively used to create artificial subsurface caries lesions in vitro .[51, 53, 73, 84, 123, 124] In such solutions, also cont aining varying concentrations of calcium, phosphate and fluorine, lesion forma tion typically occurred over the course of several weeks. The addition of F to the demi neralization solution encourages a form of remineralization through the formation of a mineral-rich surface layer above a scarcelymineralized caries lesion zone. Ofte n rich in F, this remineralized layer inhibits the subsequent demineralization of the surface enamel. The dissoluti on potential of lactic acid is decreased in this solution, thus explaining the long treatment time for caries lesion formation. The focus of this research is on the surface effects of enamel dissolution; extensive demineralization times are not needed. In this study, as-received human incisor enam el slabs were immersed in 0.12M EDTA and buffered lactic acid to illicit enamel surface di ssolution. Effective topographical changes were examined using AFM and SEM. Crystallinity studies were performed using the conventional -2 XRD and GIXD techniques to an alyze the bulk and demineralized surface layers of enamel, respectively. Nanoindentation compared hardness a nd elastic moduli data to that calculated from as-received enamel. Raman data, correlated with confocal imaging, provided relative chemical information within the demineralized enamel surface layers.

PAGE 77

77 Materials and Methods Enamel Demineralization Treatments Select samples evaluated in Chapter 4 were immersed in 0.12M EDTA for 10, 30 and 60 min time intervals as outlined in Chapter 3. Th e remaining as-received samples were immersed in a buffered lactic acid solution for 30 min at human body (37C) and room (20C) temperatures. The treated samples were characte rized as outlined in Ch apter 3, unless otherwise indicated. Demineralized Enamel Characterization Two-dimensional AFM images and represen tative roughness (RMS) values of enamel treated with EDTA and buffered lactic acid were generated. Amplitude signal images, representative of the topographical cha nges experienced by samples, were produced. Additionally, height signal images of the sa me regions mapped surface depth information. A flattening function was performed on all height-s ignaling images to remove unwanted artifacts prior to roughness analyses. Additional SEM images of enamel treated wi th EDTA were generated. Demineralization patterns in surface enamel as a result of acid treatment time were explored. High resolution x-ray scattering was performed on enamel treated with EDTA and buffered lactic acid. In some cases, conventional -2 XRD scans were generated for comparison purposes. Special emphasis was placed on evaluating the dimensional changes in enamel crystallites upon demineralization in the c-axis direction. Nanomechanical studies were performed on enam el treated with EDTA and buffered lactic acid as outlined in Chapter 4. These results were intended to highlight the structural stability of surface enamel subjected to varying degrees of demineralization.

PAGE 78

78 Confocal imaging was performed on select enamel samples treated with EDTA and buffered lactic acid as outlined in Chapter 4 to confirm microstructural changes observed by AFM/SEM. These images, in combination with Raman spectra generated from them, provided information regarding the surface chemistry of the analyzed enamel layers. Results Demineralization 0.12M EDTA Figure 5-1 shows AFM images of the tr ansformation of as-received enamel upon demineralization with 0.12M ED TA for 30 min. Texturing of the enamel was seen in the amplitude signal imagea typical indication of su rface demineralization. The height differences depicted by the height-signal im age suggest that the linear impr ints along the sample surface, likely produced by polishing, dominat e any noticeable depth changes that may have occurred upon dissolution. An RMS roughness value of 315 nm was calculated from the textured enamel area. The same transformation of as-received to EDTA-demineralized enamel is shown by a secondary electron FE-SEM image in Figure 52. An increase in surface roughness, and the nonuniform emergence of enamel rod outlines was seen. Additional FE-SEM images highlighting the enamel dissolution extremes chosen for this study are shown in Figures 5-3 and 5-4. After a treatment time of 10 min, the nonuniform emergence of enamel rod outlines was seen. Large areas of polished enamel remaine d, indicating a slight degree of dissolution. Conversely, highly-roughened surf ace enamel observed after a 60 min EDTA treatment showed preferential demineralization of the peripheries of the enamel rods, with voided areas intermittently dispersed along the enamel surface. In early dissolution experiments, enamel slabs were immersed in small volumes of solutionless than 25 mlduring treatment. Upon visu al inspection, a decrease in intensity of the (002) reflection and increase in that of the (112) reflection was observed by GIXD ( = 0.8)

PAGE 79

79 after a 60 min EDTA treatment (Figure 5-5). The em ergence of the (211) reflection was seen as a shoulder on the (112) reflection, a peak typically resolved in bulk enamel XRD spectra. The latter occurrence was not seen during the relati vely short, 10 min tr eatment. In a single experiment, crystallite size in the [002] direction increase d from 25.56 nm to 28.71 nm as the EDTA treatment time increased from 10 min to 60 min. These results sugge sted a change in the near-surface enamel crystal structure under th e aforementioned experimental conditions, an observation with a direct influence upon subseq uent experimental de sign. As mechanical property and topographical evalua tions of these samples were not performed, the potential resistance of apatite dissoluti on after the 60 min treatment time could not be verified. In an attempt to evaluate enamel dissolution effects onl y, future experimental conditions were adjusted such that a greater volume of treatment solution was used. A subsequent dissolution experiment on a representative EDTA-treated sample showed GIXD spectra depth profile comparisons ( = 0.4) of as-received enamel and that treated for 30 and 60 min. A progressive slight sharpening of the enamel peaks with treatment time was observed (Figure 5-6). The emergence of the (211 ) reflection after the 60 min treatment was not detected. More obvious progressive improvements in crystallinity were seen with the (112) reflections. Figures 5-7 shows an averaged depth profile of changes in enamel crystallite size from the surface toward bulk enamel. An increas e in crystallite size w ith treatment time was observed closest to the enamel surface (Figure 5-8) The greatest changes in crystallite size were calculated with the 60 min treatment closest to the enamel surface at = 0.15. There was a positive correlation (R2 = 0.9541) between enamel crystallite size and ED TA treatment duration per the conditions of this experiment (Figure 5-9). The crysta llite sizes appeared to converge to a smaller value range closer to bulk enamel, suggesting that de mineralization had negligible

PAGE 80

80 effects on crystallinity in this region, or that the effects c ould not be resolved by XRD methods. Figure 5-10 shows a downward shift of the enam el XRR curve upon a 30 min EDTA treatment, suggesting a decrease in ma ss density along the surface. Results of nanoindentation testing for EDTA-tre ated enamel, as compared to as-received enamel, are displayed in Table 5-1. Comparable nanomechanical property values were calculated for as-received enamel and that treated for 10 mi n, although the latter show ed a slight advantage (Figures 5-11 and 5-12). Repres entative load vs. displacement curves are shown in Figure 5-13. After the 60 min treatment, the en amel showed a sharp decrease in nanohardness with values along the same order of magnitude as that for den tin. The stiffness of dentin is calculated to be higher than that of the 60 min demineralized en amel, as evidenced by the discrepancy in elastic modulus values. A graphical repr esentation of the nanomechanical behavior of enamel treated with EDTA for 60 min and den tin is shown in Figure 5-14. Confocal Raman studies on enamel treated with EDTA for 60 min showed a highly roughened surface, highlighting the i rregular detection of enamel ap atite (Figure 5-15). Brighter areas, color-coded as orange, on the X-Y image represented areas of higher apatite concentration. The X-Z depth profile showed barely-distinguishable outlines of enamel rods, with brighter apatitic areas along the enamel surface. The averaged Raman spectrum indicated a relative decrease in phosphate band intensity and an overa ll noisy spectrum (Figure 5-16). The presence of amorphous carbon peaks in the 1200/cm to 1850/cm range indicated laser burning of the sample. The C-H stretching band present in the as-received Raman spectrum appeared to be overshadowed by noise in the spectrum produced by demineralized enamel. Black regions of the confocal images likely reflect the presence of organic material and/or underfocused areas.

PAGE 81

81 Demineralization Buffered Lactic Acid Figure 5-17 shows AFM images of as-received enamel upon demineralization with buffered lactic acid at 20C. Small patches of seemingly-punctured enam el, indicating potential demineralization as evidenced by mineral loss in th ese regions, were seen. Evidence of polishing lines disappeared in the amplitude-signal image, but their presence was detected in the heightsignal image. An RMS roughness value of 28.4 nm was calculated. Treated enamel at 37C showed less evidence of microstruc tural transformation, as seen in Figure 5-18. In addition to the surface debris detected, faint polishing lines were still visible. An RMS roughness value of 14.1 nm was calculated. SEM images of an additiona l room temperature treatment of enamel are shown in Figure 5-19. Backscatter imaging was us ed in addition to secondary electron imaging to show better contrast between the surface features. Uniform pitting was observed; no polishing lines were discerned. There was a slight curvature to some of the pits, potentially indicating the beginning stages of enamel rod dissolution. Trea tment duration for all of the representative images shown was 30 min. A GIXD-derived depth profile comparison of as-received enamel and that treated for 30 min at 37C shows a slight decrease in crysta llite size closest to the enamel surface at = 0.15, although there was quite a bit of overlap throughout the plot (Figure 5-20). Conversely, a representative GIXD spectra comparison of as-r eceived enamel and that treated at 37C ( = 0.4) in Figure 5-21 shows an overall decrease in peak intensity. The (200) reflection also disappeared after treatment. Similar crystallinity results were observed upon treatment of asreceived enamel at 20C. Figure 522 indicates a slight decrease in crystallite size with treatment at = 0.15. A comparison of as-received and buffere d lactic acid-treated XRR curves in Figure 523 shows negligible change in the c for the enamel, but a downwar d shift in the lower portion of the curve. This indicates an increase in roughness upon treatment.

PAGE 82

82 Results of nanoindentation testing for enamel treated with buffered l actic acid at 37C and 20C, as compared to as-received enamel, are displayed in Table 5-1. Comparable nanomechanical property values were calculated for both treatment temperatures (Figures 5-24 and 5-25). Representative load vs. displacement curves are shown in Figure 5-26. Interestingly, significant decreases in nanohar dness and elastic moduli were observed upon treatment. These values appear to contradict the minimal demine ralization effects seen with AFM/SEM imaging. Confocal Raman studies on enamel treated wi th buffered lactic acid at 20C showed a rather smooth surface, similar to that seen for as-received enamel in Section 4.3 (Figure 5-27). Red color-coded areas on the X-Y image represent apatite. The averaged Raman spectrum indicated the presence of sharp 14 phosphate bands, but at lower relative intensities than those seen for as-received enamel (Figure 5-28). Ne gligible organic material was detected. Discussion Enamel demineralization tends to occur in areas most susceptible to dissolution. These regions may vary locally along the enamel surf ace due to impurity concentration gradients and the presence of micros tructural defects, causi ng selective enamel disso lution. This may explain the laterally inhomogeneous di ssolution pattern seen produced during the shorter EDTA treatment times. The highly-roughened surf ace produced upon prolonged EDTA treatment verified the preferential dissolution of enamel near the enamel rod peripheries, a phenomenon widely reported in the literature.[80, 106, 125] This is different than the etch patterns produced by other etchants, such as phosphoric and certain con centrations of citric acids, where the formation of craters along the central regions of enamel rods has been reported.[31, 52] A possible explanation, reported by Johnson et al. ,[106] suggested that the large size of the EDTA molecule restricts it from easily penetrati ng the enamel structure except for regions between enamel rods and/or where crystallite orientat ion is not parallel to the c-ax is enamel rod direction. Buffered

PAGE 83

83 lactic acid dissolution effects were relatively mild, with few areas of minor demineralization along the enamel surface. Enamel rod dissolution app eared to be in its earliest stages after a 30 min treatment, as opposed to instances of full peripheral demineralizati on with EDTA after 10 min. As such, 0.12M EDTA is a more aggressive demineralization agent. The buffering effect of the lactic acid solution was realized by the lack of texturing of the surface layer, particularly evident after the body temperatur e treatment. Dissolution, although s light and represented by pits along the enamel surface, appeared to be more a ggressive at room temperature. As buffered acid solutions are commonly used to create caries lesions in enamel by artificial means, a demineralization agent that encourages s ubsurface lesion development beneath a wellmineralized surface layer is desirable. As such, initial enamel surface pro tection is expected. Based solely upon AFM image comparisons, a hi gher treatment temperature may decrease the extent of buffered lactic acid enamel de mineralization. In earlier experimental studies, the enamel sl abs were immersed in less than 25 ml of EDTA for solution conservation purposes. This wa s just enough to cover the surfaces of the enamel slabs. The unexpected emergence of th e (211) reflection afte r the 60 min treatment suggested a possible enamel struct ural change at the surface upon demineralization. As stated in Chapter 4, texturing properties and the positi oning of the sample within the diffractometer prevented this reflection from being resolved in GIXD studies of as-received enamel. It is assumed that enamel dissolution through the removal of Ca2+ and PO4 3caused an increase in ion concentration in solution so great that it reached a supersaturated state, encouraging redeposition of the apatite onto the enamel surface. An increase in intensity of (112) suggests an increase in the number of properly-oriented crystal grains available for diffraction, another indication of possible redeposition or crystall ite growth along the [112] direc tion. The calculate d increase in

PAGE 84

84 crystallite size in the c-axis, [ 002], direction also supports this conclusion. It is interesting note that redeposition did not follow the crystal orie ntation pattern observe d at the surface, but showed greater similarity to that seen in bulk enamel. Redeposition of enamel crystallites during demineralization has been reporte d, and is described in terms of the self-inhibiting dissolution behavior of HAP.[126, 127] This idea was based on the concep t of constant composition solution techniques used for enamel remineralization studies.[128] A form of the constant composition technique called flow-through, or high volume, contro ls the saturation levels of important ions in solution necessary for remineralization.[123] Unbalanced ion concentrations in solution do not occur, which may cause unfavorab le shifts in the calcium and phosphate supersaturation levels required for remineralization. The remineralization process will not cease prematurely. Based on this concept, it was hypothesized that undersaturation of the tr eatment solution will encourage demineralization. Once the solution becomes saturated with HAP structural ions, demineralization will cease. In tuitively, enamel dissolution e ffects observed through AFM and SEM imaging would translate to a decrease in crys tallinity of the apatite mineral. These effects were initially thought to be masked by this oc currence of enamel redeposition/growth. As such, larger volumes of solution were used to mo re closely simulate the constant-composition experimental design. EDTAand buffered lactic acid-based e xperiments produced c ontradictory results regarding enamel surface crystallinity. Enamel progressively treated with EDTA showed an increase in crystallite size in the c-axis dir ection as a function of treatment time. This is counterintuitive, as acid attack is known to degrade the mineral structure of enamel. In an individual crystallite, enamel dissolution begins along the c-axis of the central core region. A dark line in this region, thought to represent a di fferent form of apatite or a planar defect, was

PAGE 85

85 observed in TEM studies as mentioned in Chapte r 2. As dissolution contin ues, the crystallite hollows, leaving a tubular shell of apatite.[129] Further dissolution causes fracturing of these hollow tubes into mere fragments of apatite. In this study, perhaps smaller-sized enamel crystallites that have experienced some degree of dissolution previously had completely dissolved after treatment. When performing XRD experiments, the diminishing effects of these smaller crystallites would not be a contributing factor in peak geometry. As such, the XRD peaks would not exhibit the broadening e ffects resulting from the presence of smaller crystals. Such an occurrence may have basis in the Gibbs-Thompson[130] effect for nanoscale particle dissolution and has been previously reported as an e ffect of the acid demineralization of bone.[131] Enamel dissolution kinetics may also play a role in th e observed increase in crystallinity. Crystallite surface diffusion occurring during the enamel di ssolution process may encourage the bonding, or fusion, of neighboring cr ystallites, potentially cr eating larger regions of planar alignment. This phenomenon has been observed in HR-TEM enamel caries lesion development and enamel crystal growth studies.[39, 71] Since XRD crystallinity is detected as crystal grains of similar lattice orientation, it would be enhanced in this case. That is, larger crystall ites would be detected by XRD. It is important to note th at crystallite bonding does not lead to perfectly-aligned lattices by the individual crystals. Defect s such as edge dislocations and small boundary grain boundaries can be expected from such a process. The liter ature has cited instances of lattice mismatch between bonded crystallites on the order of 1-2 a negligible consideration for XRD detection (Figure 5-29). XRR studies showed a decrease in density and increas e in roughness upon EDTA demineralization. A decrease in enamel crystallite size closes t to the surface in the buffered lactic acid experiments may reflect a dissolution condition synonymous to that which has been observed on

PAGE 86

86 the micron scale. The convergence of calculate d crystallite sizes a nd the great overlap experienced in the bulk suggests th at the acid solutions only affect ed the upper surface layers in this study. The selected composition of the lactic acid solution illicits demineralization that can take several days or weeks to produce artificial caries lesions, and may take a similar amount of time for discernable demineralization indicators to appear on the enamel surface. XRR curve comparisons suggest a negligible change in mineral density upon demineralization, yet an increase in surface roughness. Based on the nanoindentation resu lts observed in this study, en amel hardness and stiffness is compromised with extent of EDTA demi neralization. Extensive EDTA demineralization yields enamel deformation closest to that of dentin in this stu dy, although dentin appears to be more resilient. Subsurface demine ralization of the buffered lactic acid-treated samples may have occurred, accounting for limited enamel surface roughness observed by AFM and SEM. The increase in mechanical property values impl y that demineralization has occurred, although enamel integrity appears to be at a relatively optimal level under human body demineralization conditions as seen in Figure 5-26. The increase in enamel crystallite size on the nanoscale does not appear to increase the overall hardness or strength of enamel. This effect may be due to nanoindentation probe placement and/or the inho mogeneous nature of demineralized enamel. Erroneous data may be produced from areas where the probe cannot achie ve adequate contact with the sample surface. Sample roughness will also limit the accuracy of nanoindentation results.[18, 33, 132] While more consistent results were achieved with as-received enamel, there is higher data variability when greater surf ace irregularities are introduced. As such, nanoindentation results of treat ed enamel are better discussed in relative terms.

PAGE 87

87 Confocal Raman microscopy showed extensive de struction of the enam el structure after a 60 min EDTA treatment, which correlates with a major decrease in hardness calculated by nanoindentation. Relatively bright areas of apatite along the upper por tions of select enamel rods may reflect areas of greater mineral content. Th is may support the increase in crystallite size observed through GIXD crystallite calculations upon demineraliza tion. As the Raman technique is based on laser light-scattering effects, the debilitating refractive properties of the highlydemineralized enamel produced a rather no isy spectrum. The amorphous carbon by-product may have been generated by the intera ction of the laser with organic material within the tooth, which may have been easily exposed to the laser due to the extensiv e damage done to the enamel apatite structure. In the case of the buffered lactic acid-treated enam el, negligible organic material was detected along the sample surface. This further supports the suggested formation of an apatitic surface la yer upon treatment. Enamel dissolution is a complicated process that includes a combination of extensive topographical dissolution at the micron level, a d ecrease in mechanical integrity on the nanoscale and some structural improvement at the crystallin e level. As seen in this research, assumptions about improved crystallinity, and its enhancing effects on the stru ctural integrity of surface enamel, cannot be made relying solely upon cha nges in mineral structure seen on the micron scale. Demineralization agent sele ction and enamel treatment time greatly affects the extent of enamel damage, an important consideration in tooth decay treatment option therapy. Lastly, stringent experimental conditions and precise calibration and ope ration of the characterization equipment are necessary to provide accurate as sessments of mineral ch anges that occur during early enamel dissolution.

PAGE 88

88 Subsequent remineralization evaluations perf ormed by this research will focus on apatite regeneration effects on EDTA-treated enamel, due to the greater degree of material property changes observed upon demineralization. Examples of the remineralization of buffered lactic acid-treated enamel substrates will be included for comparison purposes.

PAGE 89

89 Figure 51. AFM images highlightin g transformation of as-received en amel upon 0.12M EDTA demineralization for 30 min. Scale = 10 m2 Figure 52. FE-SEM image highlighting transf ormation of as-received enamel upon 0.12M EDTA demineralization for 30 min.

PAGE 90

90 Figure 53. FE-SEM image of as-received enamel treated with 0.12M EDTA for 10 min. Figure 54. FE-SEM image of as-received enamel treated with 0.12M EDTA for 60 min.

PAGE 91

91 Figure 55. Representative GIXD spectra comparison of as-received enamel treated with < 25 ml of 0.12M EDTA solution at = 0.8. A) 10 min treatment. B) 60 min treatment. Figure 56. Representative GIXD spectra compar ison of as-received and EDTA-treated enamel at = 0.4. A) As-received. B) 0.12M EDTA treatment for 30 min. C) 0.12M EDTA treatment for 60 min.

PAGE 92

92 15 20 25 30 35 40 45 0.150.20.250.30.40.81.0Crystallite Size (nm)Grazing Incidence Angle, () As-Received 10 min 30 min 60 min Figure 57. Depth profile of the average enamel crystallite size along the caxis as a function of for EDTA enamel treatment regime. 10 20 30 40 50 60 0.10.150.20.250.30.350.40.45Crystallite Size (nm)Grazing Incidence Angle, () As-Received 10 min 30 min 60 min Figure 58. Depth profile of th e average surface enamel crystall ite sizes along the c-axis as a function of for EDTA enamel treatment regime. Error bars represent 95% confidence.

PAGE 93

93 R = 0.9541 0 10 20 30 40 50 60 70 202325283033353840Treatment Time (min)Crystallite Size (nm) Figure 59. Correlation of EDTA treatment time with crystallite size at = 0.15. 1.0E+01 1.0E+02 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.20.50.71.01.21.51.72.0log Intensity (a.u.)Theta As-Received EDTA-30 min Figure 510. Representative XRR curve comparison of as-received and EDTA-treated (30 min) enamel.

PAGE 94

94 Table 5-1. Comparison of as-received and demine ralized enamel hardness and modulus values. Treatment Hardness (GPa) Elastic Modulus (GPa) As-Received 4.86 0.44 95.87 5.58 As-Received Dentin* 0.59 0.26 21.08 6.44 0.12M EDTA 10 min 4.87 1.26 102.63 18.87 0.12M EDTA 60 min 0.28 0.10 4.82 1.25 Lactic Acid 37C 1.65 0.36 73.27 8.87 Lactic Acid 20C 1.74 0.87 63.04 18.08 *Included for comparison purposes only 0 1 2 3 4 5 6 As-Received10 min EDTA60 min EDTA Figure 511. Comparison of as-received and EDTA-treated enamel nanohardness values. Error bars represent 95% confidence.

PAGE 95

95 0 20 40 60 80 100 120 140 As-Received10 min EDTA60 min EDTA Figure 512. Comparison of as-received and EDTA -treated enamel elastic modulus values. Error bars represent 95% confidence. Figure 513. Comparison of representative forc e (load)-displacement curves for as-received and EDTA-treated enamel. A) As-received. B) 10 min EDTA treatment. C) 60 min EDTA treatment.

PAGE 96

96 Figure 514. Comparison of representative force (load)-displacement curves for EDTA-treated enamel and as-received dentin. A) 60 min EDTA treatment. B) As-received dentin. Figure 515. Confocal microscopy images of as -received enamel treated with 0.12M EDTA for 60 min. A) X-Y surface scan. B) X-Z cross-sectional scan.

PAGE 97

97 Figure 516. Averaged Raman spectrum of as -received enamel treated with 0.12M EDTA for 60 min. Note th e prominent amorphous carbon bands due to laser burning of the sample. Figure 517. AFM images of as-received enamel tr eated with buffered lactic acid for 30 min at room temperature. Scale = 20 m2

PAGE 98

98 Figure 518. AFM images of as-received enamel tr eated with buffered lactic acid for 30 min at human body temperature. Scale = 10 m2 Figure 519. SEM images of as-received enamel tr eated with buffered lactic acid for 30 min at room temperature. Scale bar = 20 m. A) Secondary electron image. B) Backscattered electron image.

PAGE 99

99 10 15 20 25 30 35 40 45 50 0.150.20.250.30.40.81.0T-2TCrystallite Size (nm)Grazing Incidence Angle, ( ) As-Received 30 min -Body Temp Figure 520. Depth profile of the average enamel crystallite size along the c-axis as a function of for a 30 min buffered lactic acid treatment at human body temperature. Figure 521. Figure 5-20. Representative GIXD sp ectra comparison of as-received and buffered lactic acid-treated enamel at = 0.4. A) As-received. B) Buffered lactic acid treatment for 30 min at human body temperature.

PAGE 100

100 10 15 20 25 30 35 40 45 50 0.150.20.250.30.40.81.0T-2TCrystallite Size (nm)Grazing Incidence Angle, () As-Received 30 min -Room Temp Figure 522. Depth profile of the average enamel crystallite size along the c-axis as a function of for a 30 min buffered lactic acid treatment at room temperature. 1.0E+01 1.0E+02 1.0E+03 1.0E+04 1.0E+05 1.0E+06 0.20.50.71.01.21.51.72.0log Intensity (a.u.)Theta As-Received 30 min -Room Temp Figure 523. Representative XRR curve comparison of as-received enamel and treated with buffered lactic acid for 30 min at room temperature.

PAGE 101

101 0 1 2 3 4 5 6 As-ReceivedLactic Acid -37CLactic Acid -20C Figure 524. Comparison of as-received and buffered lactic acid-treated enamel nanohardness values. Error bars repr esent 95% confidence. 0 20 40 60 80 100 120 As-ReceivedLactic Acid -37CLactic Acid -20C Figure 525. Comparison of as-received and EDTA -treated enamel elastic modulus values. Error bars represent 95% confidence.

PAGE 102

102 Figure 526. Comparison of representative forc e (load)-displacement curves for as-received and 30 min buffered lactic acid-treated en amel. A) As-received. B) Human body temperature treatment. C) Room temperature treatment. Figure 527. Confocal microscopy X-Y surface image of as-received enamel treated with buffered lactic acid for 30 min at room temperature.

PAGE 103

103 Figure 528. Averaged Raman spectrum of enamel treated with buffered lactic acid for 30 min at room temperature. Note absence of organic material bands within 2000/cm to 2500/cm range. Figure 529. TEM image showing the bonding of ne ighboring crystallites w ithin a caries lesion. The white arrow indicates the central dark li ne. As the crystal grows toward the left, it begins to fuse (black arrow) with a ne ighboring crystal. Latti ces of the original crystal (broken line) are 1 out of alignment with lattic es of the neighboring crystal (solid line).[39]

PAGE 104

104 CHAPTER 6 ENAMEL REMINERALIZATION Introduction The lack of scientific evidence to complement observations of enamel remineralization has been a subject of controversy among research ers. As stated in Section 2.4, the process of remineralization has been determined to ensue in one of three ways: regeneration of partia lly-dissolved crystals emergence of newly-formed crystals additional growth of existing crystals The first of these mechanisms is most commonly referred to as remineralization in the literature. Although numerous radiographic studies involving subsurface caries lesions have reported the (re)generation of apatite upon treatment,[73] there is uncertainty surrounding how this process occurs. Has enamel mineral regrow th occurred or has a compositionally-different apatite phase been precipitated? Solution-mediat ed apatite phase transformations occur and tend to be influenced by solution pH, temperature, ion concentration, and co mposition. For example, dicalcium phosphate dihydrate (DCPD) can precipitate onto apatitic surfaces during demineralization in phosphate-rich acidic solu tions (pH 3-6) at temperatures of 37C.[133] This is most relevant in caries lesion development. At higher solution pH values, octacalcium phosphate (OCD) forms at a pH of 6.8 to 7.0, reprecipitated apatite forms at a pH of 7.5 to 11.0. Enamel impurity concentrations may also have an eff ect on apatite phase transformation. For example, amorphous calcium phosphate may tran sform to DCPD at pH values eq ual to or greater than 7 in the presence of Mg2+.[63] Microhardness studies have cited an increase in hardness of enamel with remineralization, garnering increased s upport for the theory of enamel mineral rebuilding.[91, 134-136] It is assumed that an increase in the mechanical properties directly correlates with an increase in the structural integrity of tooth enamel. While there is extensive data

PAGE 105

105 supporting the occurrence of remineralization as measured on the micron scale, there is little evidence that this phenomenon occurs on the submicron scale. When discussing remineralization, the infl uence of fluoride cannot be ignored. Low concentrations of F added to calciumand phosphate-rich remineralization solutions, on the subppm scale, have been reported to en courage enamel mineral regeneration.[137] Other studies have shown that the inclusion of 0.1 ppm F in soluti on during remineralizati on has prompted calcium uptake, but no enhancing effects of F on demineralized enamel were detected.[138] The remineralization of the inner subsurface caries lesion zone and outer surface layer was reported at F concentrations within the range of 2 to10 ppm.[81] TEM imaging studies have proven that an increase in enamel crystallite size occurs when subjected to the aforementioned solution types, and this effect is claimed to be influenced by the presence of F.[41] The filling of centrallyhollowed demineralized crystall ites with apatite upon reminerali zation was also demonstrated, suggesting mineral enhancement on the submicron level.[71] These results have been corroborated by XRD studies, which cite an increase in crystallite c-axis le ngth and a-axis width dimensions.[57] While F does appear to have an enhancing effect on the regeneration of enamel mineral, is it a necessary component or does it act as a catalyst for the remineralization process? In this study, selected EDTA-treated enamel slabs were subjected to MI Paste and a synthetic calcifying solution (0.1 ppm F) to ill icit enamel surface reconstruction. Resultant surface enamel microstructural changes were ex amined by AFM and SEM. Crystallinity studies were performed using GIXD techniques to analyze the demineralized surface layers of enamel. Nanoindentation provided a relative comparison of hardness and elastic moduli data relative to the data calculated from EDTA-demineralized enamel. Raman data, corre late with confocal imaging observations, and provide relative chem ical information within the remineralized

PAGE 106

106 enamel surface layers. Selected buffered lactic acid -treated enamel slabs were remineralized with MI Paste and the calcifying solution for microstructural and limited crystallinity comparison purposes only. Materials and Methods Enamel Remineralization Treatments Selected EDTA-treated samples evaluated in Chapter 5 were subjecte d to MI Paste for 5and 10-min time intervals as outlined in Chapter 3. Other EDTA-treated samples were immersed in a calcifying solution for 8 h. For comparison purposes, selected buffered lacticacid-treated samples were similarly remineralize d. Characterization was performed as outlined in Chapters 3 and 4. Since limited data were ge nerated for the buffered lactic acid substrate samples, the statistical significance of those results were not confirmed in this study. Remineralized Enamel Characterization Two-dimensional AFM images and representative roughness (RMS ) values of the remineralized enamel samples were generated. Topographical comparisons of the remineralized surface enamel, as a result of remineralizing agent type, were made. Secondary electron SEM images of remineralized enamel samples were generated. Remineralization effects as a result of remine ralization treatment type were explored. High resolution x-ray scattering was performed on remineralized enamel samples. Special emphasis was placed on discerning the dimens ional changes in enamel crystallites upon demineralization in the c-axis direction. Nanomechanical studies were performed on re mineralized enamel samples. These results were intended to highlight the structural stability of surface enamel subjected to differing sources of remineralization.

PAGE 107

107 Confocal imaging was performed on selected remineralized enamel samples to confirm microstructural changes observed by AFM/SEM. These images, in combination with Raman spectra extracted from color-coded enamel phase s, provided information regarding the chemistry within the enamel surface layers. Results MI Paste Treatments EDTA-treated enamel substrate Figure 6-1 shows AFM images of demineralized enamel (0.12M EDTA for 30 min) after a 5-min treatment with MI Paste. Irregularly-shaped, raised patches were dispersed uniformly along the surface, with little eviden ce of the extent of demineralization observed in Figure 5-1. An averaged RMS roughness value of 29.2 nm was calculated over th e entire image. A secondary electron FE-SEM image of a similarlyprepared sample is shown in Figure 6-2. The enamel rod outlines are more defined, with a gr eater accumulation of material within the center of the enamel rods and along their peripheries (Figure 6-3). An image focusing on the central region of a single enamel rod shows bundled regions of what appears to be enamel crystallites (Figure 6-4). A representative GIXD spectra depth profile ( = 0.4) of enamel treatments, culminating with a 5-min MI Paste application, showed a pronounced decrease in crystallinity along the crystallite c-axis direction after remineralizati on (Figure 6-5). This is clearly depicted by the (002) relative peak intens ity distribution in Figure 6-6. A second 5-min application of the paste did little to enhance crystallin ity. No new phases, indicated by the emergence of new diffraction peaks, were seen. A depth profile graph of crys tallite size changes from the surface toward bulk enamel in Figure 6-7 shows little evidence of impr oved crystallinity with remineralization in the [002] direction. A decrease in crystallite size wa s detected on the scale of that observed after a

PAGE 108

108 10-min EDTA treatment of as-received enamel (Figure 6-8). A surpri sing consequence of remineralization was the enhanced crystallinity depicted by the (200) and, to a lesser degree, (111) reflections shown in Figure 6-5. Evidence of increased (200) peak intensity/narrowing with remineralization is highlighted in Figure 6-9, which was enhanced with a second application of paste. In terms of the stability of reminera lized enamel treated by MI Paste, an average increase in enamel crystallinity in the c-axis di rection of 5 nm was determined after immersion of the remineralized enamel slab in distilled water for 24 h (Figure 6-10). The results depicted in this figure are based on the GIXD analysis of a single enamel slab. Results of nanoindentation testing for the MI Paste remineralization potential of EDTAtreated enamel, as compared to previously-r eported as-received and demineralized enamel values, are summarized in Tabl e 6-1. Nanohardness results are comp arable to those calculated for 60-min EDTA-treated enamel, although the fo rmer showed a slight advantage. There is overlap in the calculated values, making it difficu lt to propose significant differences in enamel hardness (Figure 6-11). The averag e elastic modulus value for the remineralized enamel was an order of magnitude higher than that of the ED TA-treated enamel (Figure 6-12). Representative load vs. displacement curves are shown in Figure 6-13. Confocal Raman results were extracted from th e region enclosed in the red square from the video image of textured enamel shown in Part A of Figure 6-14. Confocal imaging of this area produced an X-Y scan in Part B of the figur e, highlighting patches of apatite (red) among potential residue from the past e (blue) and a combination of both (pink). The lower region burned by the laser was omitted as indicated. A de pth profile of the sample surface layers in Figure 6-15 showed sporadic bright red areas of apatite within the paste residue. Black regions within the confocal images likely reflect unde rfocused sample surface areas. The averaged

PAGE 109

109 Raman spectrum extracted from the X-Y scan showed rather sharp 1 4 phosphate bands in the lower spectral region, in addition to bands indicative of organic material at the higher wavenumber value ranges (Figure 6-16). The C-H stretching band present in the as-received Raman spectrum, which may be shifted in the EDTA-treated Raman spectra, was replaced by a stronger organic material presence. This is indi cated by the relatively, well-defined peaks toward the higher end of the spectrum. Buffered lactic acid-treated enamel substrate AFM images of a 5-min MI Paste treatment on buffered lactic-acid-demineralized enamel in Figure 6-17 show granular patches of matter dispersed ra ndomly along the sample surface. Enamel rod outlines were faintly det ected in some areas. An RMS roughness value of 99.4 nm was calculated over the entire image, perhaps primarily due to the two large humps observed on the surface. Figure 6-18 shows a decrease in crystallite size in the [002] direction closest to the surface, with a marked increase in crystallinity at de pths greater than 120 nm. Calcifying Solution Treatments EDTA-treated enamel substrate Figure 6-19 shows AFM images of demineraliz ed enamel (0.12M EDTA for 30 min) after an 8 h treatment with the calcifying solution. The demineralized enamel microstructure was still fairly visible, with raised pa tches of matter randomly distribute d along the surface. An averaged RMS roughness value of 408 nm was cal culated over the entire image. A representative GIXD spectra comparison ( = 0.4) of EDTA-dem ineralized enamel immersed for 8 h in a calcifying solution showed su ch an overall decrease in peak intensity such that the peaks could barely be discerned (Figure 6-20). There was also a decrease in peak intensity within the halo region. A representative depth profile graph of crystallite size changes from the surface toward bulk enamel in Figure 6-21 showed a negligible change in crystallinity

PAGE 110

110 with remineralization in the [002] direction be yond that originally detected for as-received enamel. The effects of the MI Paste treatment a ppeared to be negated. Th is behavior ensued to depths of approximately 228 nm. Results of nanoindentation testing for the ca lcifying solution remine ralization of EDTAtreated enamel, as compared to previously-r eported as-received and demineralized enamel values, are displayed in Table 6-1. Average na nohardness was significantly higher than that calculated for the 60-min, EDTA-treated enam el. The average elastic modulus for the remineralized enamel was an order of magnitude higher than that for the EDTA-treated enamel, although statistically lower than that calculated after MI Paste remineralization. Representative load vs. displacement curves are shown in Figure 6-13. Buffered lactic acid-treated enamel substrate AFM images of an 8-h calcifying solution trea tment on buffered lactic acid-demineralized enamel in Figure 6-22 showed a sample surface of variable height, with irregularly-shaped globules of lightly-textured matter randomly disp ersed. Enamel surface height was less uniform than in samples evaluated elsewhere for this research. An RMS roughness value of 20 nm was calculated over the entire image, a value closer to that calculated for buffered lactic acid-treated enamel at human body temperature. Figure 6-23 shows random aggregation of enamel crystallites similar to that seen in EDTA-demineralized and MI Paste-remineralized enamel. A representative increase in crystallite size in the [002] directi on closest to the surface was seen in Figure 6-24; this behavior dropped sharply as surface enamel transitioned into bulk. Discussion MI Paste Treatments Upon remineralization with MI Paste, AFM imaging suggests that a reaction between the demineralized enamel and the paste occurred sporadically on the surf aces of both EDTAand

PAGE 111

111 buffered lactic acid-treated samples. This is represented by patches of textured enamel not observed through pure demineralization. Through visual observation it is unclear whether the surface layer consists of precip itates, enamel that has undergon e a phase transformation or a combination of both. In FE-SEM images of the ED TA-treated substrate, MI Paste appears to have enhanced the sizes and/or accumulation of enamel crystallite s in the central and peripheral regions of the enamel rods. One particular claim as to the functionality of the paste is the ability to bind calcium and phosphate to dental surfaces.[139] Crystallite agglomeration may have occurred as a result of the adhesion forces be tween the paste and apatite. A previously-done FE-SEM study of remineralization using a dilu ted CPP-ACP paste reported rather smooth enamel surface appearance upon subsequent demineralization, barring faint enamel rod outlines.[140] This supports the theory that MI Past e discourages demineralization, perhaps through the additional buffering action of an actual surface layer that adheres to dental tissue. A remnant of this layer is perhaps seen on the buffered lactic acid subs trate. Structural and chemical analyses would provide insight into the degree of crystallinity (v s. amorphous nature) present along the remineralized surfaces, as well as identify any new phases formed. A representative GIXD spectra comparison of an MI Paste-treated sample on an EDTAdemineralized substrate showed a decrease in crysta llinity in the apatite c-axis direction. This can be attributed to a dominating surface layer that may be partly amorphous in nature, limiting GIXD detection of crystalline apatite along th e (002) plane. A comp lementary GIXD depth profile of crystallite size changes from the su rface toward the bulk showed the potential for increased crystallinity (on average) from an as-rece ived enamel state, but the sample size was too small to make a statistically sound assessment. Of particular inte rest was the decrease in enamel crystallinity depicted by the ( 200) reflection upon EDTA deminer alization and the subsequent

PAGE 112

112 increase in crystallinity upon MI Paste re mineralization shown by a representative sample analyzed in Figures 6-6 and 6-9. The (200) plane belongs to the fa mily, of which the (300) plane (2 = 32.9)[117] is a member. The latter is the reflection typically used in -2 XRD analyses to determine HAP crystallite size in the a-axis direction.[38, 141-143] As such, dimensional changes in the [200] direction may correlate to enamel cr ystallite changes along th e a-axis direction. As stated in Chapter 4, the (200) re flection is typically not resolved in conventional XRD analyses. Its presence in GIXD spectra may compensate fo r this loss in assessing a-axis crystallite dimensions along the surface. Also, widening of the enamel crystallite as a result of MI Paste remineralization may cause a decrease in the crystallite length dimension, assuming constant crystallite volume. Nevertheless, preferential growth along the (200) plane seems to occur upon MI Paste remineralization. The peaks from eac h representative spectrum in Figure 6-5 appear well-aligned with respect to their 2 positions, suggesting that a phase transformation was unlikely. Another interesting occurrenc e was the apparent enhancement of the MI Paste remineralization capabilities upon im mersion in distilled water. Th e fluid appears to serve as a catalyst in driving the ion di ffusion processes necessary for potential enamel regeneration. Improved crystallinity on the orde r of 5 nm was observed in the [002] direction closest to the surface after immersion. This suggests promise for the interaction of the paste with saliva, in vivo in producing enhanced enamel crystallinity results when in contact with fluid. Nanoindentation studies detected, at best, a minimal improvement in remineralized enamel hardness due to the processes explored in this study. This finding contradicts the improvements in enamel microhardness reported in the l iterature by MI Paste remineralization of demineralized enamel.[91] Another study of remineralization vi a a calcifying solution (5 ppm F)

PAGE 113

113 suggested that enamel rehardening did not o ccur based on microhardness studies.[58] Instead, intercrystalline spaces in enam el were filled with finely-g rained, amorphous precipitates upon remineralization to improve hardness of the enamel tissue, collectively. This may actually account for improvements in mineral vol% obs erved by microradiography studies claiming enamel remineralization. Extending this rationale to MI Paste remineralization effects, the accumulation of crystallites observed in enamel rod core regions and around their peripheries may be mechanically stable. Intercrystalline space is present between enamel rods in the form of the enamel organic matrix. Mineral dissolution in the core regions is often accompanied by an increase in intercrystalline space. Nanoindent er probing of such areas would yield lower hardness results, but improvements in enamel stiffness due to the increased presence of apatite material in these regions. Based on nanoinde ntation curve comparisons in Figure 6-13, remineralization improves the resiliency of the enamel examined in this research. Chemical analyses performed within the surface layers of remineralized enamel by confocal Raman microscopy identified the patches of textured material as apatite. Unfortunately, the representative apatitic phase could not be di scerned due to instrument al limitations. Confocal Raman analyses also detected the presence of organic material along the surface, likely due to the complex carbohydrate and polymer-bas ed components found in MI Paste.[144] Likewise, Raman could not discern the precise chemical composition of the detected organic material. Energy dispersive spectroscopy (E DS) has the ability to extract relative concentrations of key compositional elements to aid in stoichiometr y determination, and would prove useful in identifying the phases present as a result of enamel remineralization. Analysis of the MI Paste remineralizati on of a buffered lactic acid-demineralized substrate showed little effect on crystallinity at the enamel surface, with possible measurement

PAGE 114

114 errors in deeper layers showing improved crystal linity in the crystallite [002] direction. Due to a lack of statistically-relevant crystallinity and nano indentation data, the true effects of MI Paste remineralization cannot be verified. Calcifying Solution Treatments Upon remineralization with a ca lcifying solution (0.1 ppm F), a slight surface layer seemed to form on both EDTAand buffered lactic acid-d emineralized enamel samples. This assumption was verified for the EDTA-treated substrate th rough a GIXD spectra comparison near the surface of remineralized enamel highlighting the obvi ous decrease in peak, and amorphous halo, intensities. For EDTA-demineralized enamel, a decrease in crystallite size below that of asreceived enamel was detected by GIXD. These re sults suggest that an amorphous surface layer may have obstructed pertinent peak detection by GIXD. For buffered lactic acid-demineralized enamel, an increase in crystallite size along th e enamel surface was seen. This may be due to compound effects influenced by F on the relatively well-mineralized substrate produced by the buffered lactic acid (0.5 ppm F) demineralization, although it is interesting to note that there appeared to be little change in crystallite size based on SEM observation between the MI Pasteand calcifying solution-treated enamel. Of course, in such an instance, the underlying demineralization effects on crysta llite size cannot be ignored. Enamel remineralization appears to be a complicated process that depends on a variety of factors, namely enamel F concentration, the st ate of demineralized enamel and experimental environment/conditions. In evaluating remine ralization effects through changes in XRD crystallinity, a crystallite size comparison of as -received and remineralized enamel may be most appropriate in drawing conclusi ons about the integrity of toot h enamel upon remineralization, as this study cannot confirm the validity of the structural e ffects seen as a result of demineralization. A more focused study utilizing a greater sample size, ut ilizing instrumentation

PAGE 115

115 with the ability to conduct elemental analyses at the enamel surface, would provide a better indication of structural mechanisms behind enamel remineralization.

PAGE 116

116 Figure 61. AFM images of EDTA-t reated enamel (30 min) subjected to MI Paste for 5 min. Scale = 20 m2 Figure 62. FE-SEM image of ED TA-treated enamel (30 min) s ubjected to MI Paste for 5 min.

PAGE 117

117 Figure 63. FE-SEM image of ED TA-treated enamel (30 min) s ubjected to MI Paste for 5 min, highlighting a single enamel rod. Figure 64. FE-SEM image of ED TA-treated enamel (30 min) s ubjected to MI Paste for 5 min, highlighting central region of a single enamel rod.

PAGE 118

118 Figure 65. Representative GIXD spectra comparison of as-received, EDTAand MI Pastetreated enamel at = 0.4. A) As-received. B) 0.12M EDTA treatment for 30 min. C) EDTA + MI Paste treatment for 5 min. 0 100 200 300 400 500 600 2525.52626.5Intensity (a.u.)Theta As-Received 30 min EDTA EDTA+ MI Paste (5 min) EDTA + MI Paste (10 min) Figure 66. (002) GIXD spectra comparison for as-received, EDTA-treated (30 min) and MI Paste-treated (5 and 10 min) enamel at = 0.4.

PAGE 119

119 15 20 25 30 35 40 45 50 0.150.20.250.30.40.81.0Crystallite Size (nm)Grazing Incidence Angle, () As-Received 10 min 30 min 60 min MI Paste -5 min Figure 67. Depth profile of the average enamel crystallite size along the caxis as a function of for as-received enamel progressively treated with EDTA, e nding with a 5 min MI Paste treatment. 15 20 25 30 35 40 45 50 55 0.10.150.20.250.30.35Crystallite Size (nm)Grazing Incidence Angle, () As-Received 10 min 30 min 60 min MI Paste -5 min Figure 68. Depth profile of th e average surface enamel crystall ite sizes along the c-axis as a function of for as-received enamel progressively treated with ED TA, ending with a 5 min MI Paste treatment. Erro r bars represent 95% confidence.

PAGE 120

120 300 500 700 900 1100 20.52121.52222.5Intensity (a.u.)Theta As-Received 30 min EDTA EDTA+ MI Paste (5 min) EDTA + MI Paste (10 min) Figure 69. (002) GIXD spectra comparison for as-received, EDTA-treated (30 min) and MI Paste-treated (5 and 10 min) enamel at = 0.4. 0 10 20 30 40 50 60 70 80 0.150.20.250.30.40.81.0Crystallite Size (nm)Grazing Incidence Angle, () As-Received 60 min MI Paste -0 h MI Paste -24 h Figure 610. Representative depth profile of the average enamel crystallite size along the c-axis as a function of for as-received enamel treated with EDTA for 60 min, ending with a 5 min MI Paste treatment. Treatment stability highlighted upon immersion in distilled water for 24 h.

PAGE 121

121 Table 6-1. Comparison of hardness and elastic modulus values of as-received, EDTAdemineralized and enamel remineralized w ith MI Paste and a calcifying solution (0.1 ppm F). Treatment Hardness (GPa) Elastic Modulus (GPa) As-Received 4.9 0.4 95.9 5.6 0.12M EDTA 60 min 0.3 0.1 4.8 1.3 0.12M EDTA + MI Paste 0.2 0.2 19.1 13.5 0.12M EDTA + Calcifying Soln 0.4 0.3 12.6 8.1 0.0 0.1 0.2 0.3 0.4 0.5 60 min EDTAEDTA + 5 min MI Paste EDTA + 8 h Calcifying Soln Figure 611. Comparison of EDTA-treated and subsequently remineralized enamel nanohardne ss values. Error bars represent 95% confidence.

PAGE 122

122 0 5 10 15 20 25 60 min EDTAEDTA + 5 min MI Paste EDTA + 8 h Calcifying Soln Figure 612. Comparison of EDTA-treated and subsequently remineralized enamel elastic m odulus values. Error bars represent 95% confidence. Figure 613. Comparison of representative forc e (load)-displacement curves for as-received, 60 min EDTA-treated and remineralized enamel. A) As-received. B) 60 min EDTA treatment. C) EDTA + 8 h calcifyi ng soln treatment. D) EDTA + 5 min MI Paste treatment.

PAGE 123

123 Figure 614. X-Y surface images of 60 min EDTA + MI Paste-treated (5 min) enamel. A) Video image; Scale bar = 10 m. B) Confocal microscopy image; Scale bar = 6 m Figure 615. Confocal microscope X-Z cross-sectional image of 60 min EDTA + MI Pastetreated (5 min) enamel. Scale bar = 6 m

PAGE 124

124 Figure 616. Averaged Raman spectrum of 60 min EDTA + MI Paste-treat ed (5 min) enamel. Note heavy presence of organic ma terial bands within 1250/cm to 3500/cm range. Figure 617. AFM images of buffered lactic acid -treated enamel (room temperature) subjected to MI Paste for 5 min. Scale = 20 m2

PAGE 125

125 20 22 24 26 28 30 32 34 36 0.150.20.250.30.40.81.0T-2TCrystallite Size (nm)Grazing Incidence Angle, () As-Received 30 min -Room Temp MI Paste (5 min) Figure 618. Representative depth profile of the average enamel crystallite size along the c-axis as a function of for as-received and buffered lactic acid-treated enamel, ending with a 5 min MI Paste treatment. Figure 619. AFM images of 60 min EDTA-treated enamel subjected to a calcifying solution for 8 h. Scale = 20 m2

PAGE 126

126 Figure 620. GIXD spectra comparison of as-re ceived, EDTAand calcifying solution-treated enamel at = 0.4. A) As-received. B) 0.12M EDTA treatment for 30 min. C) Calcifying soln treatment for 8 h. 0 10 20 30 40 50 60 70 80 0.150.20.250.30.40.81.0Crystallite Size (nm)Grazing Incidence Angle, () As-Received 60 min MI Paste -5 min Calcifying Soln -8 h Figure 621. Representative depth profile of the average enamel crystallite size along the c-axis as a function of for as-received and 60 min EDTA-treated enamel, ending with a 5 min MI Paste and a 8 h calcifying solution treatment.

PAGE 127

127 Figure 622. AFM images of buffered lactic acid -treated enamel (room temperature) subjected to a calcifying solution for 8 h. Scale = 20 m2 Figure 623. FE-SEM image of 30 min buffered l actic acid-treated enamel (room temperature) subjected to a calcifying solution for 8 h, highlighting the central region of a single enamel rod.

PAGE 128

128 20 22 24 26 28 30 32 34 36 0.150.20.250.30.40.81.0T-2TCrystallite Size (nm)Grazing Incidence Angle, () As-Received 30 min -Room Temp Calcifying Soln -8 h Figure 624. Representative depth profile of the average enamel crystallite size along the c-axis as a function of for as-received and 30 min buffered lactic acid-treated enamel, ending with an 8 h calcifying solution treatment.

PAGE 129

129 CHAPTER 7 CONCLUSIONS Important Findings from the Research The purpose of this study was to determine stru ctural changes within the surface layers of tooth enamel upon demineralization and remi neralization on the submicron scale. GIXD provided a depth profiling of crystallinity ch anges from the surface to several hundreds of nanometers into the enamel, a resoluti on only achieved through TEM methods. The nanoindentation process was refine d for as-received enamel to achieve results comparable to those cited in the literature. The nanomechanical behavior of treated enamel was compared to these baseline values to determine relative mechan ical effects. These results were corroborated with well-established morphological imaging and Raman scattering techniques in hopes of providing greater insight into the mechanisms su rrounding early stages of enamel dissolution and regeneration. Similarly, GIXD was used to exam ine the nature of the material produced along surface enamel as a result of remineralization. This study also aimed to determine whether epitaxial regrowth of apatite occurred on part ially-demineralized enamel, as proposed by a number of researchers based on microrad iography, microhardne ss and conventional -2 XRD studies. Enamel slabs, cut from intact human inciso r teeth, were highlypolished to a relative mirror-like finish for standardization purposes throughout this study. Through GIXD experiments, surface enamel was found to have a slightly different cr ystal structure along its surface. The emergence of the (200) peak, which is unable to be discerned through conventional XRD methods, was evident. Upon the treatmen t of surface enamel with 0.12M EDTA, SEM imaging showed decomposed regions of distinct enamel rod outlines randomly dispersed along an otherwise smooth enamel surface after 10 min which progressed into extensive dissolution

PAGE 130

130 near the periphery of the enamel rod walls af ter 60 min. Confocal Raman imaging and spectral changes after a 60-min EDTA treatment confir med major enamel demineralization. On the crystalline level, GIXD detected an increase in crystallinity upon 0.12M EDTA treatment most evident to approximately 1000 nm below the enamel surface. This counterintuitive result may be a function of the dissolution of smaller enamel crystallites upon demineralization, or the bonded of neighboring crystallites as a result of surf ace diffusion processes. Nanoindentation studies showed appreciable difference in hardness and el astic modulus after an extensive 60-min EDTA treatment, with values shifting from 4.86 0.44 GPa for as-received enamel to 0.28 0.10 GPa and from 102.63 18.87 GPa to 4.82 1.25 GPa, respectively. Although errors may convolute the true effects of the treatments on the mechan ical properties calculated, these relative changes appear reasonable. Collective resu lts suggest that mineral dissol ution evident on the micron scale may be governed by an increase in enamel crys tallinity on the nanoscale, leading to extensive decomposition and highly increased fragility of tooth enamel. While an intuitive decrease in enamel crystallinity was observed with buffered lactic acid-treated samples, demineralization was too slow to adequately quantify the enamel property changes seen. The remineralization of extens ively-demineralized enamel substrates showed preferential growth along the (200) plane (crystallite aaxis dimension) upon treatment with a CPP-ACP topical paste. This did not translate to an incr ease in nanohardness of the enamel, as its average value remained relatively constant at 0.2 0.2 GPa. The elastic modulus increased to 19.1 13.5 GPa, suggesting an increase in m echanical resiliency upon remineralization. Structural effects of th e calcifying solution (0.5 ppm F) were inconclusive as a surface layer appeared to form on the enamel substrate. While remineralization studies with the topical paste and calcifying solution were al so conducted on buffered lactic acid-demineralized enamel

PAGE 131

131 substrates, it was difficult to determine significant changes seen due to the limited demineralization effects experienced by the samples. Although most of the results from this resear ch cannot be substantiated through statistical means due to limitations in samples sizes for each treatment and other resources, many collective sets of data showed anticipated trends that ma y be considered in future studies of the surface structural effects of demineralization and remineralization. Future Work General trends and variability in calculated enamel crystallinity, hardness and modulus suggests that the enamel dissolution and reminera lization are much more complicated processes than originally thought. In order to substantiate the findings of this research, additi onal studies in the area of surface crystallinity are needed. Studies conducted on a larger scale would help in verifying trends in such a variable experimental substrate as tooth enam el. Dissolution studies incorporating a constant-composition experimental design would ensure that adequate solution ion concentration levels are maintained in order better control the demineralization process. HR-TEM studies would provide in formation on the true action of enamel crystallites subject to EDTA demineralization, namely whether cr ystallite bonding does truly occur. The cross-sectional examination of enamel surf ace layers with the TEM may be achieved upon improvements in sample preparation methods rea lized by the FIB. The brittle nature of asreceived enamel, and potential hi gh reactivity of treated enamel with the ion beam, makes this technique a difficult one. If samp le preparation is achieved, a de pth profiling of the enamel electron density changes upon treatm ent can be correlated with surf ace x-ray scattering studies. High-resolution EDS elemental analysis would also provide insight into possible apatite phase changes so slight that they are difficult to disc ern with XRD studies alone Finally, a refinement of the nanoindentation testing process for treate d enamel may prove useful in accounting for

PAGE 132

132 greatly enhanced surface roughness and potential pliability of th e enamel surface. Further down the line, the introduction of environmental fact ors affecting enamel di ssolution and regeneration to better simulate intraoral conditions would prove useful.

PAGE 133

133 APPENDIX A ENAMEL POLISHING Introduction To aid in the standardization of enamel sample surfaces prior to characterization, a polishing study was performed. Fluoridated enamel is typically found within the uppermost 100 m of enamel measured from the physical surf aces of the tooth. The incorporation of F into the apatite crystal structure produces a more stable enamel region of improved crystallinity. Since the concentration of F is highly dependent on the oral care practices and/or consumption of fluoridated water by an individu al, its presence in enamel tends to vary greatly from tooth-to-tooth. Discerning the potential F e ffects on enamel crystallinity, morphology and mechanical properties is a challenging task. Rem oval of the fluoridated apatite layer would leave behind a more suitable substrate for subs equent surface enamel evaluations. Materials and Methods Seven epoxied tooth samples were successively polished with SiC abrasive papers, of a 240 grit size range, to expose highly-polished, en amel surfaces. In this study, SiC papers of select grit sizes were chosen based on anticip ated enamel removal rela tive to the abrasive paper particle sizes. The samples were attached to a hand-held disc grinde r to aid in producing level, flat surfaces. The samples were polished on a Polimet polisher (Buehler Ltd., Evanston, IL) at a wheel speed setting of 5 (moderate). Ea ch sample was polished by applying moderate pressure for 5 s at each grit size. After polis hing at each grit size, the sample was rotated 90 to ensure removal of any previ ously-made scratche s or debris. The epoxied samples were intermittently monitored for amount of enamel loss using a penetrometer (Lab-Line Instruments, Melrose Park, IL). The instrument probe was lightly positioned in the central region of each enamel surface, effectively measuring the relative heights

PAGE 134

134 of the sample. A smaller height measurement was an indication of enamel loss. Baseline measurements were taken after minimal polishing of the samples with a combination of 180 and 240 grit papers. The best resolution available fr om the penetrometer was 0.1 mm. The samples were also visually inspected prior to each measurement to check for exposure of the underlying dentin tissue. Prior to characterization, the enamel slabs were additionally subjected to an alumina slurry polishing regimen as outlined in Chapter 3. It wa s determined that enamel loss due to slurry polishing was too small to be detected by the pe netrometer. As such, those measurements were omitted from this study. Results Relative sample thicknesses of epoxied samples are listed in Table A-1. An average enamel loss of 0.4 0.075 mm was calculated. Upon visual inspecti on, all samples appeared to expose enamel only at the surface. Discussion As the average mineral loss detected was much greater than 100 m (0.1 mm), this polishing regimen seems suitable for the removal of fluoridated enamel from the sample surface. In addition to F content, overall enamel thic kness varies from tooth-to-tooth. The possible exposure of dentin is a concern, and must be mo nitored carefully. In addition to visual surface inspections, cross-sectional exam inations of the remaining enamel thickness of each sample should be considered in this research.

PAGE 135

135 Table A-1. Epoxied tooth sample thickness values after polishing at speci fied SiC abrasive grit sizes. Sample Thickness (mm) SiC Grit Size 1 2 3 4 5 6 7 180/240 5.2 6.1 7.2 5.4 5.3 4.0 3.9 320/400 5.1 6.0 6.9 5.1 5.1 3.7 3.5 600/1200 4.9 5.8 6.9 5.0 5.0 3.6 3.4 Overall Loss (mm) 0.3 0.3 0.4 0.4 0.3 0.4 0.5

PAGE 136

136 LIST OF REFERENCES [1] U. S. Dept of Health and Human Services, Oral health in americ a: a report of the surgeon general National Institute of Dental an d Cranialofacial Research, National Institutes of Health, Rockville, MD 2000. [2] U. of Texas Health Sci Center at S. Antonio, in ScienceDaily 2008. [3] B. T. Amaechi, S. M. Higham, J Dent 2005, 33, 243. [4] J. D. Bader, D. A. Shugars, J Evid Base Dent Pract 2006, 6, 91. [5] C. Murdoch-Kinch, M. McLean, J Am Dent Assoc 2003, 134, 87. [6] J. Bader, D. Shugars, A. Bonito, J Public Health Dent 2002, 62, 2001. [7] R. Tang, L. Wang, C. Orme, T. Bonstein, P. Bush, G. Nancollas, Angew Chem Int Ed 2004, 116, 2751. [8] B. Gillings, M. Buonocore, J Dent Res 1961, 40, 105. [9] A. R. ten Cate, Oral Histology: Development, Structure and Function The C. V. Mosby Company, St. Louis 1989 [10] N. Harris, F. Garcia-Godoy, Primary Preventative Dentistry Appleton & Lange, Stamford 1999. [11] M. Mathew, S. Takagi, J Res Natl Inst Stand Technol 2001, 106, 1035. [12] L. Moss-Salentijn, M. Hendericks-Klyvert, Dental and Oral Tissu es An Introduction Lea & Febiger, Philadelphia 1990. [13] T. Aoba, Anat Rec 1996, 245, 208. [14] O. Fejerskov, E. Kidd, Dental Caries: The Disease and its Clinical Management Blackwell Munksgaard, Oxford 2003. [15] F. C. M. Driessans, Mineral Aspects of Dentistry Vol. 10, Kargar, Basel, Switzerland 1982. [16] T. Attin, W. Buchalla, M. Gollner, E. Hellwig, Caries Res 2000, 34, 48. [17] M. Eisenburger, J. Hughes, N. X. West, K. D. Jandt, M. Addy, Caries Res 2000, 34, 289. [18] M. Finke, J. A. Hughes, D. M. Parker, K. D. Jandt, Surf Sci 2001, 491, 456. [19] J. t. C. Arends, J M, J Cryst Growth 1981, 53, 135.

PAGE 137

137 [20] A. K. Johansson, R. Sorvari, D. Birkhed, J. H. Meurman, J Dent 2001, 29 333. [21] P. E. Benson, S. M. Higham, N. Pender, Orthod Craniofacial Res 2003, 6, 242. [22] J. Thewlis, Proc R Soc Med 1940, 33 387. [23] A. F. Hall, J. P. Sadler, R. Strang, E. de Josselin de Jong, R. H. Foye, S. L. Creanor, Adv Dent Res 1997, 11, 420. [24] B. Angmar, P. Carlstrom, J. E. Glas, J Ultrastruct Res 1963, 8, 12. [25] J. Arends, J. J. ten Bosch, J Dent Res 1992, 71, 924. [26] C. Ganss, J. Klimek, N. Schwarz, Arch Oral Biol 2000, 45, 897. [27] J. Rees, T. Loyn, B. Chadwick, J Dent 2007, 35, 355. [28] T. Attin, K. Weiss, K. Becker, W. Buchalla, A. Wiegand, Oral Diseases 2005, 11, 7. [29] T. A. Koulourides, in Advances In Oral Biology (Ed: P. Staple), Academic Press, Inc., New York 1966, 149. [30] W. C. Oliver, G. M. Pharr, J Mater Res 1992, 7, 1564. [31] M. E. Barbour, D. M. Parker, G. C. Allen, K. D. Jandt, Eur J Oral Sci 2003, 111, 258. [32] S. Habelitz, S. J. Marshall, G. W. Marshall, M. Balooch, Arch Oral Biol 2001, 46, 173. [33] F. Lippert, D. M. Parker, K. D. Jandt, Caries Res 2004, 38, 464. [34] P. Dutta, Current Sci 2000, 78, 1478. [35] B. D. Cullity, S. R. Stock, Elements of X-Ray Diffraction Prentice-Hall, Upper Saddle River 2001. [36] A. D. Krawitz, Introduction to diffraction in ma terials science and engineering John Wiley & Sons, New York 2001. [37] H. Gawda, L. Sekowski, H. Trebacz, in WCU Paris 2003. [38] T. Sakae, Arch Oral Biol 1988, 33, 707. [39] T. Yanagisawa, Y. Miake, J Electron Micros 2003 52, 605. [40] A. F. Marshall, K. R. Lawless, J Dent Res 1981, 60, 1773.

PAGE 138

138 [41] J. W. Simmelink, S. C. Abrigo, Adv Dent Res 1989, 3, 241. [42] E. F. Bres, J. L. Hutchison, J Biomed Mater Res 2002, 63, 433. [43] S. Takagi, H. Liao, L. C. Chow, Caries Res 2000, 34, 281. [44] D. T. Zero, "Dentrifices, mouthwashe s, and remineralization/caries arrestment strategies", presented at Biotechnology and Biom aterials to Reduce the Caries Epidemic Seattle, 2005 [45] B. Ogaard, Caries Res 2001, 35, 40. [46] H. Duschner, H. Gotz, B. Ogaard, Eur J Oral Sci 1997, 105, 466. [47] H. Tsuda, W. L. Jongebloed, I. Stokroos, J. Arends, Caries Res 1993, 27, 445. [48] J. M. ten Cate, J Dent Res 1990, 69, 614. [49] C. Robinson, J. Kirkham, A. C. Baverstock, R. C. Shore, Caries Res 1992, 26, 14. [50] R. M. O. Argenta, C. P. M. Tabchoury, J. A. Cury, Pesqui Odontol Bras 2003, 17, 241. [51] J. M. ten Cate, P. P. E. Duijsters, Caries Res 1982, 16, 201. [52] B. T. Amaechi, S. M. Higham, Arch Oral Biol 2001, 46, 697. [53] L. M. Silverstone, J. S. Wefel, J Cryst Growth 1981, 53, 148. [54] K. Collys, R. Cleymaet, D. Coomans, Y. Michotte, D. Slop, Caries Res 1993, 27, 15. [55] F. Lippert, D. M. Parker, K. D. Jandt, J Colloid Interface Sci 2004, 280, 442. [56] T. Aoba, Y. Moriwaki, Y. Doi, M. Okazaki, J. Takahashi, T. Yagi, J Oral Pathol Med 1981, 10, 32. [57] A. Vieria, R. Hancock, H. Limeback, M. Schwartz, M. Grynpas, J Dent Res 2003, 82, 909. [58] H. R. Muhlemann, H. Lenz, K. Rossinsky, Helv Odont Acta 1964, 8, 108. [59] K. Kawasaki, Y. Tanaka, O. Takagi, Arch Oral Biol 2000, 45, 797. [60] R. Z. Wang, S. Weiner, J Biomech 1997, 31, 135. [61] F. H. Jones, Surface Science Reports 2001, 42, 75.

PAGE 139

139 [62] F. C. M. Driessans, R. H. M. Verbeeck, Eds., Biominerals, CRC Press, Boca Raton 1990. [63] R. Z. LeGeros, Calcium phosphates in oral biology & medicine S Kargar AG, Basel, Switzerland 1991. [64] R. H. Selwitz, A. I. Ismail, N. B. Pitts, Lancet 2007, 369, 51. [65] M. Takahashi, J. Zheng, K. Mori, I. Mataga, K. Kobayashi, Materials Science and Eng C 2006, 26, 703. [66] H. Komatsu, H. Yamamoto, M. Namachi, K. Yasuda, Y. Matsuda, Y. Murata, T. Kijimura, H. Sano, T. Sakai, T. Kamiya, Nucl Instr and Meth in Phys Res B 2007, 260, 201. [67] M. E. Barbour, J. S. Rees, J Dent 2004, 32, 591. [68] D. M. Ebenstein, L. A. Pruitt, Nanotoday 2006, 1, 26. [69] B. Bhushan, Ed. Nanotribology and Nanomechanics An Introduction Springer-Verlag, Berlin, Germany 2005. [70] C. Nave, HyperPhysics http://hyperphysics.phy-astr.gsu. edu/hbase/quantum /bragg.html, March 2008 [71] H. Tohda, T. Yanagisawa, N. Tanaka, S. Takuma, J Electron Microsc 1990 39, 238. [72] J. M. White, J Dent Res 1994, 73, 1560. [73] Y. Iijima, O. Takagi, J. Ruben, J. Arends, Caries Res 1999, 33, 206. [74] J. M. ten Cate, Acta Odontol Scand 1999, 57, 325. [75] N. W. Johnson, Arch Oral Biol 1966 11, 1421. [76] J. H. Meurman, J. M. ten Cate, Eur J Oral Sci 1996 104, 199. [77] P. Bogra, S. Kaswan, J Indian Soc Pedo Prev Dent 2003, 21, 79. [78] K. Kawasaki, J. Ruben, I. Stokroos, O. Takagi, J. Arends, Caries Res 1999 33, 275. [79] Z. C. Cehreli, N. Altay, Angle Orthodontist 2000 70, 22. [80] J. W. Simmelink, V. K. Nygaard, D. B. Scott, Arch Oral Biol 1974, 19, 183. [81] H. Yamazaki, A. Litman, H. C. Margolis, Arch Oral Biol 2007 52, 110.

PAGE 140

140 [82] C. D. Gibbs, S. E. Atherton, E. Huntington, R. J. M. Lynch, R. M. Duckworth, Arch Oral Biol 1995, 40 879. [83] J. L. Linton, Am J Orthod Dentofacial Orthop 1996, 110, 590. [84] J. B. D. Featherstone, Caries Res 1981 15, 377. [85] B. T. Amaechi, S. M. Higham, J Dent 2001, 29, 371. [86] E. Saxegaard, F. Lagerlof, G. Rolla, Acta Odontol Scand 1988, 46, 355. [87] S. V. Dorozhkin, J Coll Int Sci 1997, 191, 489. [88] H. U. V. Gerth, T. Dammaschke, E. Schafer, H. Zuchner, Dent Mater 2007, 23, 1521. [89] E. Saxegaard, G. Rolla, Scand J Dent Res 1988, 96, 523. [90] K. Yamaguchi, M. Miyazaki, T. Takamizawa, H. Inage, B. K. Moore, J Dent 2006, 34 230. [91] D. Tantbirojn, A. Huang, M. D. Ericson, S. Poolthong, J Dent 2008, 36, 74. [92] P. Shen, F. Cai, A. Nowicki, J. Vincent, E. C. Reynolds, J Dent Res 2001, 80. [93] G. Walker, F. Kai, P. She n, C. Reynolds, B. Ward, C. Fone, J Dairy Res 2006, 73, 74. [94] M. I. Richard, T. U. Schulli, E. Wintersberger, G. Renaud, G. Bauer, Thin Solid Films 2006, 508, 213. [95] J. F. Moulin, F. Dinelli, M. Massi, C. Albonetti, R. Kshirsagar, F. Biscarini, Nucl. Instrum. Methods Phys. Res., Sect. B 2006, 246, 122. [96] K.-J. Chao, P.-H. Liu, K.-Y. Huang, C. R. Chimie 2005, 8, 727. [97] D. C. Creagh, V. Otieno-Alego, P. M. O'Neill, Radiat. Phys. Chem. 2001 61, 757. [98] I.-M. Low, J Am Ceram Soc 2004, 87, 2125. [99] C. H. Russell, "X-ray reflectivity study of exotic materials for electronic materials", presented at Materials Research Society Symposium 2002. [100] J. Daillant, A. Gibaud, Eds., X-ray and Neutron Reflectivity: Principles and Applications Springer-Verlag, Berlin 1999. [101] H. Toraya, J Appl Cryst 1986, 19, 440.

PAGE 141

141 [102] N. E. Waters, in The Mechanical Properties of Biological Materials (Eds: J. F. V. Vincent, J. D. Currey), Cambridge University Press, Cambridge 1980, 99. [103] E. A. Carter, H. G. M. Edwards, in Infrared and Raman Spectroscopy of Biological Materials, Vol. 24 (Eds: H. U. Gremlich, B. Yan), Marcel Dekker, Inc, New York 2000, 421. [104] L. Breschi, P. Gobbi, G. Mazzotti, T. H. Ellis, E. Sacher, I. Stangel, J Biomed Mater Res 1999, 46, 315. [105] M. Eisenburger, R. P. Shellis, M. Addy, Caries Res 2004, 38, 67. [106] N. W. Johnson, D. F. G. Poole, J. E. Tyler, Arch Oral Biol 1971, 16, 385. [107] C. A. Volkert, S. Busch, B. Heiland, G. Dehm, J Microsc 2004 214, 208. [108] Y. Hayashi, T. Yaguchi, K. Ito, T. Kamino, Scanning 1998, 20, 234. [109] R. J. S. Vilchis, K. Yamamoto, Am J Orthod Dentofac Orthop 2007, 131, 646. [110] K. Hoshi, S. Ejiri, W. Probst, V. Seybold, T. Kamino, T. Yaguchi, N. Yamahira, H. Ozawa, J Microsc 2001, 201, 44. [111] R. K. Nalla, A. E. Porter, C. Daraio, A. M. Minor, V. Radmilovic, E. A. Stach, A. P. Tomsia, R. O. Ritchie, Micron 2005 36, 672. [112] C. A. Volkert, A. M. Minor, MRS Bulletin 2007, 32, 390. [113] I. H & M Analytical Services, Grazing Incidence Angle Thin Film Analysis http://www.h-and-m-analytical.c om /pdfs/grazing_incidence.pdf, February 2008 [114] L. H. He, M. V. Swain, J Dent 2007, 35, 431. [115] K. H. G. Ashbee, Fundamental Principles of Fiber Reinforced Composites Technomic Publishing Company, Inc, Lancaster, PA 1993. [116] J. L. Cuy, A. B. Mann, K. J. Livi, M. F. Teaford, T. P. Weihs, Arch Oral Biol 2002 47, 281. [117] JCPDS-ICDD, JCPDS-International Cent er for Diffraction Data 1996, card #09-0432. [118] T. Y. Tsui, G. M. Pharr, J Mater Res 1999, 14, 292. [119] P. N. de Aza, F. Guitian, C. Santos, Chem Mater 1997, 9, 916. [120] S. Habelitz, G. W. Marshall, M. Balooch, S. J. Marshall, J Biomech 2002, 35, 995.

PAGE 142

142 [121] M. Kuhar, P. Cevc, M. Schara, N. Funduk, J Prosthet Dent 1997, 77, 578. [122] M. D. Francis, NY Acad Sci 1965, 131 694. [123] J. A. K. M. Buskes, J. Christoffersen, J. Arends, Caries Res 1985, 19, 490. [124] L. M. Silverstone, British Dental J 1968. [125] S. Hoffman, W. S. McEwan, C. M. Drew, J Dent Res 1969, 48, 1234. [126] L. J. Wang, R. Tang, T. Bonste in, P. Bush, G. H. Nancollas, J Dent Res 2006, 85, 359. [127] S. Mafe, J. A. Manzanares, H. Reiss, J. M. Thomann, P. Gramain, J Phys Chem 1992, 96, 861. [128] D. J. White, Adv Dent Res 1995, 9, 175. [129] M. A. Listgarten, Arch Oral Biol 1966 11, 999. [130] A. W. Adamson, A. P. Gast, Physical chemistry of surfaces Wiley-Interscience, New York 1997. [131] S. N. Danilchenko, C. Moseke, L. F. Sukhodub, B. Sulkio-Cleff, Cryst Res Technol 2004, 39, 71. [132] M. S. Bobji, S. K. Biswas, J Mater Res 1998, 13, 3227. [133] R. Z. LeGeros, J Dent Res 1990, 69 567. [134] K. Collys, R. Cleymaet, D. Coomans, D. Slop, J Dent 1991, 19, 230. [135] H. R. Muhlemann, K. Rossinsky, A. Schatt, K. G. Konig, Helv Odont Acta 1966, 10, 28. [136] H. Delvin, M. A. Bassiouny, D. Boston, J Oral Rehabil 2006 33, 26. [137] A. P. M. Jacobson, R. Strang, K. W. Stephen, Caries Res 1991 25, 230. [138] C. D. Gibbs, S. E. Atherton, E. Huntington, R. J. Lunch, R. M. Duckworth, Arch Oral Biol 1995, 40 879. [139] E. C. Reynolds, J Dent Res 1997, 76, 1587. [140] M. Oshiro, K. Yamaguchi, T. Takamizawa, H. Inage, T. Watanabe, A. Irokawa, S. Ando, M. Miyazaki, J Oral Sci 2007, 49, 115.

PAGE 143

143 [141] T. Nakano, A. Tokumura, Y. Umakoshi, S. Imazato, A. Ehara, S. Ebisu, J Mater Sci: Matl Med 2001, 12, 703. [142] M. J. Larsen, S. J. Jensen, Arch Oral Biol 1989, 34 969. [143] A. A. Baig, J. L. Fox, R. A. Young, Z. Wang, J. Hsu, W. I. Higuchi, A. Chhettry, H. Zhuang, M. Otsuka, Calcif Tissue Int 1999, 64, 437. [144] C. Rahiotis, G. Vougioklakis, J Dent 2007, 35, 695.

PAGE 144

144 BIOGRAPHICAL SKETCH Carmen Veronica Gaines was born in Buff alo, New York. She obtained her secondary educational training from the Hu tchinson Central Technical High School, with a concentration in computer technology. Upon graduation, Carmen went on to pursue a bachelor of science degree at Cornell University, majoring in biological engi neering. She was later able to put her problem solving skills to work by joining the team at Harmac Medical Products, Inc. as a product development engineer. Her interest in the biomed ical field was solidified there, and encouraged the pursuit of more formalized education and an opportunity to conduct research at the University of Florida. She obtained her Master s degree from the Department of Materials Science and Engineering in May 2005 and comple ted her Ph.D. from the same department in May 2008. Carmen hopes to continue her research pursuits in the area of artificial organ development.