• TABLE OF CONTENTS
HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction
 Polarized infrared reflection spectroscopy...
 Infrared reflection spectroscopy...
 Aqueous durability of lithia-disilicate...
 Summary and recommendations
 Reference
 Biographical sketch
 Copyright














Title: polarized infrared reflection spectroscopic characterization and aqueous durability of lithia-disilicate glass-ceramics
CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00090214/00001
 Material Information
Title: polarized infrared reflection spectroscopic characterization and aqueous durability of lithia-disilicate glass-ceramics
Series Title: polarized infrared reflection spectroscopic characterization and aqueous durability of lithia-disilicate glass-ceramics
Physical Description: Book
Creator: McCracken, Walter J.,
 Record Information
Bibliographic ID: UF00090214
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000295258
oclc - 07863801

Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
    List of Tables
        Page vi
    List of Figures
        Page vii
        Page viii
        Page ix
        Page x
    Abstract
        Page xi
        Page xii
        Page xiii
    Introduction
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
    Polarized infrared reflection spectroscopy of single crystal lithia-silicates and quartz
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
    Infrared reflection spectroscopy of polycrystalline lithia-disilicate glass-ceramic
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
    Aqueous durability of lithia-disilicate glass-ceramic
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
    Summary and recommendations
        Page 108
        Page 109
        Page 110
    Reference
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
    Biographical sketch
        Page 117
        Page 118
        Page 119
    Copyright
        Copyright
Full Text












THE POLARIZED INFRARED REFLECTION SPECTROSCOPIC
CHARACTERIZATION AND AQUEOUS DURABILITY OF
LITHIA-DISILICATE GLASS-CERAMICS












By

WALTER J. McCRACKEN


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



UNIVERSITY OF FLORIDA


1981



































To Ward and Doris McCracken,

my parents
















ACKNOWLEDGMENTS


The author thanks G. Reginald Bell for sparking my interest in

materials science while at General Motors Institute. At the University

of Florida, the author wishes to thank J. J. Hren and D. E. Clark for

their many useful and interesting discussions. To W. B. Person the

author is indebted for his endless aid in the interpretation of many

of the results of this work related to infrared spectroscopy. The

author is especially indebted to his advisor, L. L. Hench,who provided

invaluable encouragement, advice, and assistance throughout this study.

Finally, the author thanks the Air Force Office of Scientific Research

for financial support of this work.


iii

















TABLE OF CONTENTS


Page

ACKNOWLEDGMENTS ........................................... iii

LIST OF TABLES........................................... vi

LIST OF FIGURES................... ........................ vii

ABSTRACT........................................ .. ..... xi

CHAPTER

I INTRODUCTION................................ 1

General Background of Glass Ceramics............. 1
Characterization of Glass-Ceramic
Structures...................................... 2
Selection of the Li20*2Si02 System............. 9
Objectives of Research........................... 10
Significance of Research........................ 10

II POLARIZED INFRARED REFLECTION SPECTROSCOPY
OF SINGLE CRYSTAL LITHIA-SILICATES AND
QUARTZ........................................... 12

Introduction.................................. 12
Experimental............ ................ 23
Results and Discussion.......................... 36
Conclusions...................................... 53

III INFRARED REFLECTION SPECTROSCOPY OF
POLYCRYSTALLINE LITHIA-DISILICATE
GLASS-CERAMIC ................................... 55

Introduction.................................... 55
Experimental ..................................... 56
Results and Discussion........................... 57
Conclusions........................ ...... ... 60










IV AQUEOUS DURABILITY OF LITHIA-DISILICATE
GLASS-CERAMIC................................... 66

Introduction ................................... 66
Experimental ................................... 67
Results and Discussion .......................... 68
Conclusions................................... 105

V SUMMARY AND RECOMMENDATION...................... 108

REFERENCES......................... ...................... 111

BIOGRAPHICAL SKETCH ...................................... 117
















LIST OF TABLES


Table Page

I-1 Some Technical Uses of Glass-Ceramics............ 3

II-1 Vibrational Frequencies of Infrared Spectral
Peaks in the Range 1400-800 cm-1 for Quartz,
Lithia-Disilicate and Lithia-Metasilicate........ 54

IV-1 Solution Data for 33L Glass-Ceramic Corrosion
in Acid, Neutral, and Base Aqueous Environ-
ment ............................................. 95















LIST OF FIGURES


Figure Page

I-1 Schematic of dual beam infrared spectro-
photometer.......... ........................... 6

II-1 Infrared reflection spectra of quartz from
the different angles of incidence (20* + 70)
and calculated optical constants 1 and K.
Corrected spectra (e") shows difference
between reflection spectra and absorption
spectra (after Simon and McMahon)................ 15

II-2 Crystal structure projections of lithia-
disilicate (Li2Si205) (after Liebau)............. 20

II-3 The crystal structure projections of the
(001) plane for lithia-metasilicate
(Li2SiO ) and lithia-disilicate (Li2Si205)....... 22

II-4 Polarized infrared reflection spectra of the
AgBr wire grid polarizing filter showing the
effect of polarizer orientation on the base-
line intensity.................................. 25

II-5 Reflectance-condenser apparatus with polariz-
ing filter as used in the reference beam path
of the spectrophotometer for PIRRS. Illustrates
rotation of the polarizer to give two spectra
900 apart at 450 (a) and 315" (b)................ 28

II-6 Polarized infrared reflection spectra obtained
from the (001) plane of lithia-disilicate
single crystal showing the change in spectra
with rotation of the polarizer................... 30

II-7 Lithium disilicate single crystal images from
scanning electron microscope A) secondary
electron imaging mode (mag. 20X) and b) back-
scattered electron channeling mode............... 33


vii










Figure Page

11-8 Diagram of the bands from the ECP of lithia-
disilicate. The bands inside the circle
represent the ECP in Fig. 11-7.................. 35

11-9 Polarized infrared reflection spectra of a
quartz single crystal on the prism face
(1010) ............................. ..... ..... 38
-2
II-10 Oxygen tetrahedra for Si2072 unit with one
nonbridging oxygen per tetrahedra illustrat-
ing possible Si-0 bond stretching vibrational
modes........................................... 41

II-11 Polarized infrared reflection spectra of a
lithia-disilicate single crystal on the (010)
plane.............. .................. .......... 43

II-12 Polarized infrared reflection spectra of a
lithia-disilicate single crystal on the (001)
plane........................................... 45
-2
11-13 Oxygen tetrahedra for Si207-2 unit with two
nonbridging oxygens per tetrahedra illustrat-
ing possible Si-0 bond stretching infrared
vibrational modes ............................... 48

11-14 Polarized infrared reflection spectra of
lithia-metasilicate oriented crystals on a
plane parallel to the crystal growth axis
(001)........................................... 50

11-15 Polarized infrared reflection spectra of
lithia-metasilicate oriented crystals on a
plane perpendicular to the crystal growth
axis (001).................................... 52

III-1 Infrared reflection spectra of 33L-90%
crystalline glass-ceramic and polarized
infrared reflection spectra of 33L-single
crystal from all three axes--X, Y, and Z........ 59

111-2 Infrared reflection spectra of 33L-90%
crystalline and 33L-60% crystalline glass-
ceramics ............... ................ ... ... 62

III-3 Infrared reflection spectra of 33L glass
and 33L-20% crystalline glass-ceramics.......... 64


viii










Figure Page

IV-1 Scanning electron micrographs of uncorroded
33L glass-ceramic with 0, 20, 60 and 90%
crystallinity. Surface polished with 600
grit SiC paper.................................. 70

IV-2 Scanning electron micrographs of 33L glass
after 2 h and 10 h exposure in acid, neutral,
and base aqueous solution at 1000C with
SA/V = 2.0 cm 1................................. 72

IV-3 Scanning electron micrographs of 33L-20%
crystalline glass-ceramic after 2 h and 10 h
exposure in acid, neutral, and base aqueous
solutions at 100C with SA/V = 2.0 cm-1......... 74

IV-4 Scanning electron micrographs of 33L-60%
crystalline glass-ceramic after 2 h and 10 h
exposure in acid, neutral, and base aqueous
solutions at 100C with SA/V = 2.0 cm-1......... 78

IV-5 Scanning electron micrographs of 33L-90%
crystalline glass-ceramic after 2 h and 10 h
exposure in acid, neutral and base aqueous
solutions at 100C with SA/V = 2.0 cm-1......... 80

IV-6 Infrared reflection spectra of 33L glass
before and after acid, neutral, and base
aqueous solution................................ 83

IV-7 Infrared reflection spectra of 33L-20%
crystalline glass-ceramic before and after
acid, neutral, and base aqueous corrosion........ 86

IV-8 Infrared reflection spectra of 33L-60%
crystalline glass-ceramic before and after
acid, neutral, and base aqueous corrosion........ 89

IV-9 Infrared reflection spectra of 33L-90%
crystalling glass-ceramic before and after
acid, neutral, and base aqueous corrosion....... 91

IV-10 Li concentration and Si02 concentration in
corrosion solution versus time for 33L glass
and glass-ceramics in acid, neutral, and base
aqueous solutions............................... 94

IV-11 Corrosion parameter a versus time of 0.1M HC1
corrosion test 1000C for 33L glass and glass-
ceramics ....................................... 100











Figure Page

IV-12 Corrosion parameter a versus time of H20
corrosion test at 100C for 33L glass and
glass-ceramics.................................. 102

IV-13 Corrosion parameter a versus time of 0.1M
NaOH corrosion test at 1000C for 33L glass
and glass-ceramics.............................. 104

IV-14 Various modes of glass-ceramic corrosion for
33L glass-ceramics................................ 107
















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



THE POLARIZED INFRARED REFLECTION SPECTROSCOPIC
CHARACTERIZATION AND AQUEOUS DURABILITY OF
LITHIA-DISILICATE GLASS-CERAMICS

By

Walter .J. McCracken

June, 1981


Chairman: Larry L. Hench
Major Department: Materials Science and Engineering


The purpose of this study is to better understand the structure-

property-environment relationship of glass-ceramic materials. It

is necessary to characterize the structure of glass-ceramic surfaces

in order to study their changes during aqueous corrosion. Thus, the

polarized infrared reflection spectroscopy (PIRRS) technique for single

crystal silicates is developed to aid in interpretation of infrared

reflection spectroscopy (IRRS) of polycrystalline glass-ceramics.

Polarized infrared reflection spectroscopy is used to assign

atomic vibrational modes to the peaks in reflection spectra of single

crystals of quartz, lithia-disilicate, and lithia metasilicate. To

predict the possible vibrational modes in these silicates a Si207 unit

cell is used instead of the commonly used SiO4 unit cell. The PIRRS











technique is useful in separating peaks that are very close or over-

lap in single crystal samples. This is illustrated by the fact that
-1
the broad peak at 1000 cm from the polycrystalline lithia-disilicate
-1
glass-ceramic is actually the Si-ONB:X peak at 1040 cm and the
-1
Si-ONB:Y peak at 995 cm combined. The effect on the PIRRS data

of increasing the number of nonbridging oxygens per silica tetrahedra

in the silicate structure is demonstrated.

By combining the PIRRS data of single crystal lithia-disilicate

and the IRRS data of polycrystalline lithia-disilicate glass-ceramics

the glass-ceramic surface can be characterized so that changes in

the surface structure during aqueous corrosion can be monitored. For

glass-ceramics, IRRS is primarily sensitive to the dominate phase

on the surface of the sample studied. Infrared reflection spectroscopy

does not appear to be an accurate method to measure percent crystallinity.

Aqueous durability of lithia-disilicate glass-ceramics is studied

using four volume fractions of crystallinity (0%, 20%, 60%, and 90%)

and three different pH environments (acid--0.1M HC1, base-0.1M NaOH,

and neutral--H20). The crystallization significantly affects the

corrosion behavior of the lithia-disilicate glass. In general, the

least surface damage is observed on all the materials when exposed

to acidic solution and the most extensive surface alterations occur

on all the materials exposed to the basic solutions. On glass, net-

work dissolution occurs uniformly over the entire exposed surface.

On glass-ceramics, network dissolution proceeds at different rates


xii











over the surface, with the highest rate occurring on the glassy

phase. In addition to ion exchange and network dissolution, phase

boundary attack between the glass and crystalline phases and intra-

crystalline attack are important in glass-ceramic systems exposed

to basic solutions, or to static neutral solutions for long times.

The intracrystalline attack mode occurs along the micaceous layer

(010) of the lithia-disilicate crystal exposing more X-Z planes

reducing the magnitude of the Si-ONB:Y peak in the infrared data.

Phase boundary attack and intracrystalline attack do not appear to

be significant on the glass-ceramics exposed to acidic solutions.

However, the crystalline phase maintains its crystal structure after

ion exchange with the acid solutions.


xiii
















CHAPTER I
INTRODUCTION

Glass-ceramics are defined as polycrystalline solids prepared
1
by controlled crystallization of glasses. These materials form a

relatively new class of ceramics compared to porcelains, glasses,

sintered ceramics and clay bodies. Glass ceramics are used in many

high technology and high performance applications due to their unique

processibility and physical properties.



General Background of Glass-Ceramics

The history of research into glass-ceramics starts in the 18th
2
century with M. de Reaumur. He studied the devitrification (crystal-

lization) of heat treated glass bottles that were slowly cooled by

being packed in sand. The bottles had a coarse needle like crystalline

structure and lower mechanical strength than the original glass bottle.

In the 1820's, "rocklike ceramics" were made by the crystallization

of glasses. This was done on an industrial scale basis in Europe but

never reached widespread acceptance due to processing difficulties.3'4

The need for the control of the crystallization process was first

recognized by H. H. Blau in 1933. He attempted to control the

opacity of glasses via crystallization and his concepts of controlled

crystallization were widely studied in Europe in the early 1950's.1

These studies finally led S. D. Stookey at Corning Glass Works in 1956











to develop the first Pyroceram glass-ceramic objects6 These were

formed as a glass and then heat treated, using controlled nucleation

and crystallization, to produce a fine grained glass-ceramic. This

was the second critical step in the development of glass-ceramic as

a commercially viable material.

Since the late 1950's, glass-ceramic materials have become

established in a wide variety of technical uses due to their unique

physical properties associated with controlled crystallization. Many

industries have developed specific glass-ceramic materials for use

in the areas of electronics, military devices, vacuum technology,

domestics, high mechanical wear, resistance to chemical attack, and

biological compatability.17-10 Examples of such uses are listed in

Table .I-1.

The unique material properties of glass-ceramics such as chemical

durability, mechanical strength, thermal expansion coefficient, dielec-

tric constant, and electromagnetic radiation transmittance are due

to the development of a crystalline microstructure that is usually

multiphase. Thus it is necessary to characterize the structure of

each phase of the glass-ceramic in order to attempt to understand its

behavior. It is especially important to characterize the role of each

phase with respect to environmental sensitivity in order to predict

the performance of the glass-ceramic material.



Characterization of Glass-Ceramic Structures

The two most important features necessary for characterization

of the structure of a glass-ceramic material are: 1) the identification












Table I-i. Some Technical Uses of Glass-Ceramics


Applications


Compositions


Electronics


photomachineable circuit boards
capacitors
insulators


Military
randomes (radio transparent nose cones)
ceramic seals for bomb triggers


Vacuum Technology
glass to metal seals for CRT


Domestic
cooking ware (conventional and microwave)
tableware
heating surfaces


Industrial
corrosion resistant tubing and pipes
pump impellor
ball bearings and bearing surfaces
telescope mirror blanks


Biological
implants and implant coatings
dental materials


Li20-Al202-Si0 2
Li20-ZnO-PbO-SiO2
Li20*2SiO2



MgO-Al203-TiO2-Si0 2
Li20-Al203-SiO2




ZnO-B203-Si0 2



Na20-Al203-SiO2

Li20-Al203-SiO2
MgO-Al203-SiO2



Li20-CaO-Al203-SiO2
Li20-MgO-Al203-SiO2
MgO-Al203-SiO2-TiO2
Li20-Al203-SiO2



Na20-CaO2-P 20-Si2
Li20-CaO-A1203-Si0 2











of the composition and crystal structure of each phase, and 2) the

quantification of the microstructural features of the material, i.e.,

volume fraction (V ) and size of each phase. X-ray diffraction is

generally used in determining the crystal structure and composition

by comparing experimental data with standards. 1'12 X-ray diffrac-

tion has also been used to determine the volume fraction of crystal-

linity of partially crystallized glass-ceramics.13

Quantitative microscopy measurements can be obtained by common

petrographic techniques using polarized or unpolarized light micro-
14-18
scopes. Microscopic data can also be determined using scanning

electron microscopy and various etching techniques to preferentially

etch one phase from the other to provide topographical contrasts.19

Infrared reflection spectroscopy (IRRS) is a nondestructive

characterization technique that is routinely used for studies of

glasses and glass corrosion.2026 The major advantages of IRRS are

the following: a) it does not require a vacuum, high energy radiation

or electron impingement, b) sampling depth is only 0.5 P, c) the

analysis is both rapid and inexpensive and d) it can be done using

a standard dual beam infrared spectrophotometer, as shown in Fig. I-1.

Since the late 1930's infrared reflection spectroscopy has been used

to characterize the structure and atomic bonding relationship of

crystalline solids, such as coal, diamond, and napthene.27-29 Appli-

cation of the infrared method to silicates has helped to further

explain the bonding and structure of various glasses and crystalline

silicates.30-37























Fig. I-i. Schematic of dual beam infrared spectrophotometer.











REFERENCE
BEAM


INFRARED
SOURCE

SAMPLE
BEAM


POLARIZING FILTER (OPTIONAL)


<--TYPICAL DATA


WAVENUMBER (cM-1)











In considering the durability of glass-ceramics, this material

must be viewed as a multiphase system with each phase having individual

corrosion characteristics and possible unique reactions at the phase

boundaries. The multiphase microstructure can be subdivided into

two categories: 1) the crystalline phase and 2) the amorphous phases

(or the residual glass phase).

Most research into glass corrosion has assumed that the glass

surface is a homogeneous single phase structure. This research has

been reviewed in a number of articles,38-43 bibliographies,44'45 and
23
most recently in a book by Clark, Pantano, and Hench.23 The mech-

anisms presented for glass durability fall into five categories: 1)

ion leaching--the ion exchange of alkali ions from the glass with

hydrogen (or hydronium) ions of the solution, 2) total dissolution--

the breakdown of the silicate structure at the surface (generally

after leaching occurs, but not necessarily), 3) saturation and

precipitation--the concentration of species in solution reach the

saturation limit and precipitate on the glass surface, thus altering

the surface characteristics of the glass and subsequent corrosion

resistance, 4) pitting--the uneven attack of the glass surface due

to the heterogeneities or stress concentrations in the surface, and

5) weathering--intermittent exposure of the surface to water vapor

with or without condensation, resulting in ion exchange and formation

of salts on the surface.

A study of glass corrosion mechanisms can give some insight into

the durability of a multiphase system by considering each phase










individually. However, the overall durability performance of a glass-

ceramic cannot be solely determined by separate phase by phase inves-

tigations. It is essential to take the phase boundaries into considera-

tion as well.

Some durability studies have been done on multiphase systems such

as phase separated glasses, photochemical machinable glass-ceramics,

and commercial glass-ceramics. Phase separated glasses, such as the

borosilicates, can demonstrate large differences in the rate of corro-

sion between the phases. This fact is used in the manufacturing of

Vycor vitreous silica where the alkali-rich phase dissolves away in

a mild acid solution leaving the silica-rich phase intact.46 Phase

separation can also result in a difference in thermal expansion coef-

ficients for each phase causing stress accelerated attack at the phase
47
boundary. Glass-ceramics contain phases that can differ both in

composition and structure. The crystalline phases are generally more
48
dense and have lower ion mobilities than the parent glass. However

in certain compositional ranges the crystals are most easily leached.

The varying relative durability of the glass-ceramic phases have been

used to produce photochemical machinable articles. Such materials

as Corning's Code 8603 (based on the lithia silicates), are nucleated

by UV or visible light in controlled areas of the surface. They are

subsequently heat treated for crystal growth and exposed to a corrod-

ing media. The crystallized region corrodes away leaving the glass

intact with only minor surface leaching.49 In contrast many other

glass-ceramic systems show increased durability over the parent glass

system.50










The major factors influencing the overall durability for glass-

ceramics appear to be: 1) composition of the crystalline phases as

compared to the residual glass, 2) thermal expansion of crystalline

phase relative to residual glass, 3) structure and density of the

crystalline phase, 4) volume fraction of crystalline phase, and 5)

grain size of crystalline phase. There has been no previous effort

to quantify the relative effect of these factors on the durability

of glass-ceramics.



Selection of the Li20-2Si02 System

The lithia-silica glass system is the binary basis for a large

group of commercial glass-ceramics composed of lithia-alumino-silicates.

This binary system has been used in numerous fundamental studies in

the areas of: nucleation and crystallization,51 2) phase separation,52

aqueous durability,53'54 and 4) mechanical strength.55'56 The glass

composition chosen for many of the previous studies and this investi-

gation is composed of 33 mole % Li20 and 67 mole % SiO2 (Li20-2SiO2

and also referred to as 33L). This composition is stoichiometric with

the crystalline lithia-disilicate phase that is produced by controlled

nucleation and crystallization of this glass.51 Thus, the 33L glass-

ceramic provides a two-phase solid with the same composition but with

one phase amorphous and the other crystalline. This characteristic

permits the study of crystalline effects on the properties of the

material without the complexity of major compositional variables

occurring within the structure.











Objectives of Research

This research is a twofold effort to better understand the struc-

ture-property-environment interactions of glass-ceramic materials.

The first objective is to develop a technique to characterize the sur-
34
face structure of glass-ceramics. Based on earlier work of Matossi,4

Reitzel35 and MacKenzie,36 on infrared reflection absorption spectros-

copy of organic crystalline solids,275759 the polarized infrared

reflection spectroscopy (PIRRS) technique is developed herein as a

tool to study the surface of crystalline ceramics.

The second objective is to determine the aqueous corrosion mech-

anisms for a stoichiometric binary glass-ceramic material. Using the

Li20.2Si02 glass system, a study of the effect of partial crystalliza-

tion in the glass-ceramic durability is conducted to compare with the

mechanisms of attack for homogeneous glasses.54



Significance of Research

The PIRRS technique will be used to characterize the structure

of single crystal Li20.2Si02 by the assignment of structural atomic

vibrations to the reflection peaks. These assignments are then applied

to the infrared reflection spectroscopy data for polycrystalline

Li20.2Si02 glass-ceramics, where peak overlap causes peak assignments

to be a serious problem. This technique can possibly be used for sur-

face characterization of many ceramic systems if both large and single

crystals can be obtained and the crystal structure is known.











The chemical durability of glass-ceramics has been of scientific
60
interest since the development of Pyroceram Berezhoni presents

a review of some studies of the durability of several glass-ceramic

systems (including lithia-disilicate)662 in acidic and basic aqueous

solutions. In these studies the indicator of the extent of corrosion

is weight loss measurements (and a few solution ion concentration

determinations) after corrosion. The mechanisms with which corrosion

provides for multiphase glass-ceramics are not discussed. However,

micrographs of dilute HF acid corroded lithia-disilicate do show

preferential attack along some grain boundaries and overall dissolution

of the glass from the crystalline material. The model of corrosion

mechanism of glass-ceramics herein will be useful in understanding

the behavior of other multiphase ceramics with one phase being glass

in an aggressive environment. There are similar applications of this

model in liquid phase sintered ceramics and the glassy grain boundaries

in silicon nitrides.















CHAPTER II
POLARIZED INFRARED REFLECTION SPECTROSCOPY OF
SINGLE CRYSTAL LITHIA-SILICATES AND QUARTZ



Introduction

The application of infrared spectroscopy to the identification

and characterization of unknown organic compounds is well established

and widely used. Infrared spectra of organic compounds generally

exhibit sharp well-defined peaks, which are assignable to the vibra-

tions of the individual functional groups of which the molecule is

comprised. However, the infrared spectra derived from silicate com-

pounds usually have peaks that are broad and overlapping, making assign-

ments and specific identification more difficult in this case.

The frequencies at which a material absorbs (or reflects) infrared

energy depends upon the internal vibrations of atoms in the molecule

and hence its composition. Most infrared studies of silicates assume

that the structural unit is a single Si-04 tetrahedron with zero, one,

or two nonbridging oxygens per tetrahedron which correlate to the

symmetry point groups of Td, C3V, and C2V, respectively.30-3763-66

Sanders et al. related the frequency assignments to the possible vibra-
37
tional models.37 When using infrared reflection spectroscopy these

assignments have been useful in characterizing glass structure and
20-24
structural changes associated with glass corrosion. 4 In crystal-

line silicates, the ordering of the tetrahedra gives rise to additional










peaks not predicted by a single Si-0 tetrahedron model. Thus to select

the vibrational modes that correlate with the peaks in the infrared

spectra of crystalline silicate solids, one mustuse a multi-unit (multi-

tetrahedron) model as suggested by Gordon.67

Infrared reflection methods of analysis of crystalline solids

have been employed since the late 1930's by Matossi et al.3,6869

The incident radiation sets the electrons of the atoms of the materials

into oscillation and it is this reradiation that generates the reflected

beam. The spectra for silicates generated by reflection are generally

different from the absorption spectra in that the optical constants

(n and K) of the material cause shifts in the characteristic absorp-

tion peaks. Thus the location of peaks in the reflection spectra cannot

be directly compared with either the absorption peak locations or the

"corrected" (calculated) spectra from reflection data. An example

of these differences and the effect of n and K is shown in Fig. II-1

for quartz. A complete discussion of correlating reflection spectra

and absorption spectra can be found in references 65 and 70.

Reflection of infrared radiation from dielectric materials, such

as silicates, generally produces partially plane polarized light.7

Near normal incidence reflection allows infrared sampling of approxi-
72,73
mately 0.5 im into the surface of the silicate material.7273

The absorption (or reflection) of electromagnetic radiation by

a crystalline solid produces an oscillating electric moment which is

associated with a fundamental mode of vibration. In electromagnetic

radiation that is plane polarized the electric field and the magnetic

field are perpendicular. The oscillating electron moment can only


































Fig. II-i.


Infrared reflection spectra of quartz from the
different angles of incidence (200 + 70) and
calculated optical constants n and K. Corrected
spectra (e") shows difference between reflection
spectra and absorption spectra (after Simon and
McMahon).











QUARTZ BASAL PLANE
NOT POLARIZED SPECTRA
" 70" FROM NORMAL INCIDENCE

-,-.- 20" FROM NORMAL INCIDENCE
-- (CALCULATED)
........ ...... .


/

I
i
I


I-.
iV
1j
'i


1.0


0.8


0.6


0.4


0.2




10




8



6


1200

WAVENUMBER (cM-1)


I
i


K


T7,K


1400


1000


,, ,I


..- ono


Iw


m


QBQ


-2.










interact with electromagnetic radiation that has an electric field

in its direction. Thus, by using plane polarized radiation the orienta-

tion of the fundamental mode of vibration can be determined for either

single crystals or oriented polycrystalline surfaces.

Polarized infrared radiation has been used to examine the structure

of several organic crystals as early as the 1930's.74 The advantage

of using polarized infrared reflection spectroscopy (PIRRS) is twofold.

First, by identifying the characteristic vibrational direction for

a given reflection peak the mode of molecular vibration can be determined

for the given crystal structure. When the crystallographic orientation

of the crystal is known the direction of vibration of the molecules

in the incident plane, which causes the reflection peak, must agree

with the predicted direction in order for the assignment to be correct.

Secondly, when spectra peaks overlap making them difficult to identify,

the PIERS technique is a powerful method that can aid in separating

the observed peaks without complex mathematical deconvolution techniques

using computers. Each infrared peak can be characterized accordingly

to its symmetry in the crystal structure. Peaks of different symmetry

lying close together can be distinguished by analyzing the infrared

radiation using a rotating polarizing filter at two different positions

90 apart. In an ideal case, the polarizer at one position would elimi-

nate one of the peaks from the spectra allowing clear identification

of the other peak. When the polarizer is rotated to the second position,

the former peak can be identified.

For many crystals the directions of the polarization axes are

fixed by symmetry and become independent of the frequency of light.










In uniaxial crystals of the hexagonal, trigonal, and tetragonal systems,

and in biaxial crystals of the orthorhombic system, all the polarization

axes are fixed by symmetry along the crystallographic axes. For mono-

clinic crystals one axis is fixed by symmetry while the other two are

unrestricted. Thus the spectra observed parallel or perpendicular

to polarization axes are fixed by symmetry. These axes will assume

a fundamental significance, while spectra taken in other directions

can lead to erroneous interpretations.

The single crystal silicates used for this polarized infrared

reflection spectroscopy study were quartz, SiO2 (with almost symmet-

rical SiO4 tetrahedra and zero nonbridging oxygen), lithia-disilicate,

Li2Si205 (with a single nonbridging oxygen per tetrahedra) and lithia-

metasilicate, Li2SiO3 (with two nonbridging oxygen per tetrahedra).

The crystal structure of quartz, which is based on a hexagonal

lattice, was first determined by Gibbs.75 The atomic arrangement for

Si02 is as follows:


The Si atoms are in four-coordination with oxygen, and
constitute the (Si04) tetrahedron found as the basis unit of
structure in all the known polymorphs of SiO2 and in silicates
in general. In quartz each oxygen is shared with two Si atoms,
the (Si04) tetrahedron thus being linked by sharing of each of
the corner oxygen atoms to form a three-dimensional network....

The (Si04) tetrahedron in quartz is almost symmetrical, with
an Si-O distance of 1.61 A. Each oxygen has in addition to two
Si atoms six adjacent oxygen atoms at distances ranging between
2.60 and 2.67 A. The Si-0 bond is of intermediate type, roughly
half covalent and half ionic (1962, p. 227).76


The significance of this structure lies in the fact that the SiO4 tetra-

hedra have a Td symmetry point group because each of the oxygen atoms

are bonded to two Si atoms.










The schematic projection after Liebau77 in Fig. 11-2 shows that

4
the structure of lithia-disilicate consists of Si0 tetrahedra linked

together by three oxygen atoms. The fourth oxygen atom is bonded to

a lithium atom and not shared by tetrahedra. This configuration gives

lithia-disilicate a C3V symmetry point group. The low-temperature

form of Li2Si205 is monoclinic with four formula units in the unit
77
cell. However, lithia-disilicate appears to be orthorhombic with
78
three cleavages at right angles. One cleavage parallel to (010)

is micaceous, the other two are nearly perfect. It should be noted

that some current references still list Li2Si205 as being orthorhombic.

The crystal structure of lithia-metasilicate contains two oxygen

atoms per tetrahedra bonded to lithium atoms as shown in Fig. 11-3.
79
Donnay and Donnay79 describe the Li2Si03 structure as follows:


It consists of single chains of SiO4 tetrahedra, parallel
to the C axis and held together by lithium atoms. The structure
differs from that of the pyroxenes in that it has four alkali
ions per Si206 link of the chain rather than two alkaline earth
ions.... With four lithium-oxygen bonds of strength , the
Pauling rule is perfectly satisfied since each unshared oxygen
receives four such bonds. The coordination polyhedron of lithium
roughly approximates a tetrahedron (1953, p. 166).


Li2SiO3 exists in one form with an orthorhombic pseudo-hexagonal crystal

structure.

There are three steps in the PIRRS method in order to determine

the Si-0 stretching peaks and also to assign the vibrational modes

for single crystal silicates. First, determine the crystallographic

orientation of the sample with respect to the infrared beam. The

orientation of many single crystals can be determined visually, but































Fig. 11-2. Crystal structure projections of lithia-disilicate
(Li2Si205) (after Liebau).








LITHIA-DISILICATE CRYSTAL STRUCTURE


(001) PLANE


0 --
Y


1


x4
0






1


z1
0






2


1


Z
0


0 -
Y


(010) PLANE


0 -
X


(100) PLANE































Fig. 11-3. The crystal structure projections of the (001)
plane for lithia-metasilicate (Li2Si03) and
lithia-disilicate (Li2Si205).













1 A




0 1
X
x

LITHIA-METASILICATE (LI2SIO3)








2



I V







0 -- 1
Y

LITHIA-DISILICATE (L12S1205)










some require microscopic or X-ray techniques. Second, determine the

possible modes of infrared stretching vibrations in the crystal unit

cell. Finally, coordinate the polarized infrared reflection spectra

obtained from the sample surface with the possible vibrational modes

in the incident crystallographic plane.



Experimental

Apparatus

The standard infrared reflection spectroscopic technique that

has been used previously2024375254 utilizes a dual beam Perkin-

Elmer model 467 grating spectrophotometer with a combination beam

condenser-specular reflectance accessory model 186-0373. This

accessory is added to both the sample and the reference beam. For

PIRRS, the accessory was added to the sample beam with a Venetian

blind reference beam attenuator shown in Fig. I-1, in order to adjust

the baseline. The reference beam attenuator allowed for compensation

of the AgBr polarizer element in the Perkin-Elmer wire grid polarizer

accessory model 186-0243 in the sample beam. This polarizing filter
-1 80
was selected for its extended spectral range (4000-250 cm ) which

is required for use with silicates (1300-250 cm- ). Either the

polarizer or the sample may be rotated. To minimize the polarization

effects of the grating prism dispersion elements and to produce a

flatter baseline as shown in Fig. 11-4, the polarizer was oriented

at 45 (or 3150) with respect to the entrance slits of the spectro-

photometer. The orientation of the sample stage with respect to


































Fig. 11-4. Polarized infrared reflection spectra of the AgBr
wire grid polarizing filter showing the effect of
polarizer orientation on the baseline intensity.













100




80


1300 1100 900
WAVENUMBER (cM-1)










sample space polarizer was determined using an additional polarizing

grid placed on the sample stage beneath a mirror. With the sample

space polarizer set at 450, the polarizer on the stage was rotated

to produce maximum transmittence as determined by the spectrophotometer.

This located the position on the sample stage which correlates to 450

on the sample space polarizer. Two spectra were taken, one each with

the polarizer at 450 and 315, giving maximum difference in sample

polarization effects, as shown in Fig. II-5 and 11-6.


Sample Preparation

Polarized infrared reflection spectroscopy studies require the

use of either single crystal samples or highly oriented samples. A

4 in. long doubly ended single crystal of quartz provided an excellent

sample. However, the lithia-disilicate (33L) and lithia-metasilicate

(50L) samples had to be made. Glasses were prepared of each composition,

using reagent grade materials, by melting them in a covered platinum

crucible in a SiC glo-bar furnace to reheat to 13500C. Then the glasses

were furnace cooled slowly (6-8 hrs.) to room temperature. The resultant

microstructure of each solid was comprised of large crystals. The

samples were cut with a diamond watering saw and polished with SiC

paper.

The crystallographic orientation of the samples was accomplished

by visual inspection and electron beam channeling. The sides of the

quartz sample yielded the hexagonal prism face (1010). The micaceous

cleavage direction of the 33L crystals yielded the (010) plane. The

(001) plane of the 33L crystal was obtained by cutting perpendicularly

































Fig. 11-5. Reflectance-condenser apparatus with polarizing
filter as used in the reference beam path of the
spectrophotometer for PIRRS. Illustrates rota-
tion of the polarizer to give two spectra 90
apart at 45 (a) and 3150 (b).











PIRRS APPARATUS


SPECTROPHOTOMETER
SPECTROPHOTOMETER


(a)


POLARIZER 315


SPECTROPHOTOMETER
SPECTROPHOTOMETER


SOURCE


(b)


































Fig. 11-6. Polarized infrared reflection spectra obtained from
the (001) plane of lithia-disilicate single crystal
showing the change in spectra with rotation of the
polarizer.




















.J

0 -


L)
I,














1200 1000 800

WAVENUMBER (CM-I)


33L-SINGLE CRY
POLARIZER-3I54





a
-.
C.,
U-
I
I-l






-)
cc


1200 1000 800

WAVENUMBER (CM-I)










to the micaceous layer and confirmed by electron channeling, Fig. 11-7

and Fig. 11-8. The 50L sample crystallized from the outside surface

toward the center, producing crystals with their growth axes (001)

near parallel and their (100) axes and (010) axes random.

Electron beam channeling is a scanning electron microscope tech-

nique that gives the crystallographic nature of the specimen. A de-

tailed discussion of this technique can be found in references 81 and

82. The electron channeling pattern (ECP), Fig. II-7b, is a reciprocal

space image (similar to a Kikuchi pattern) of the crystallographic

planes of the specimen. The identification of the planes was done

using the band widths and the interplanar angles. The band widths

are inversely proportional to the plane spacing. The indexing of the

ECP for a lithia-disilicate crystal is shown in Fig. 11-8. The (010)

pole of the channeling pattern indicates the surface of the crystal

is the (010) plane. Also from Fig. 11-7, the cleavage planes of the

crystal are parallel to the (400) plane and the (002) plane in the

ECP.

After the crystallographic plane of the sample surface was deter-

mined, the axes of that plane were determined by rotating the sample

on the stage of the PIRRS apparatus in order to maximize the difference

between the 450 spectra and the 3150 spectra. The change in peak

intensity in relation to the position of the polarizer is referred
83
to as dichroism. The ratio of the two peak intensities is called

the dichroic ratio. The experimentally determined dichroic ratios

may differ from the sample material's actual ratio giving rise to some


































Fig. 11-7. Lithium disilicate single crystal images from
scanning electron microscope A) secondary electron
imaging mode (mag. 20X) and B) backscattered elec-
tron channeling mode.











LITHIUM DISILICATE SINGLE CRYSTAL


A) SCANNING ELECTRON MICROGRAPH (20X)


B) ELECTRON CHANNELING PATTERN


































Fig. 11-8. Diagram of the bands from the ECP of lithia-
disilicate. The bands inside the circle
represent the ECP in Fig. 11-7.












113


113


113


202


400










possible errors in the polarized infrared reflection spectra. Some

various reasons for such a difference are imperfectly oriented samples,

imperfect polarization by the polarizer, inherent polarization by the

spectrophotometer, birefringence, and the effects of beam covergence

in condensing systems.84



Results and Discussion

The infrared reflection spectra of quartz has been the subject

of various investigations,32,34,63 some using analysis by polarizers.65

The results of this study using quartz is in Fig. II-9. The spectra

were taken from a prism face parallel to the optic axis (0001). The

two PIRRS spectra were taken at 90 to each other showing no relative

polarization.

Comparing the reflection data in Fig. II-1 (after Simon 82) and
-1 -1
Fig. II-9 shows similar spectra with two peaks at 1180 cm and 1110 cm .

As a result of the difference between reflection spectra and absorption
85 -1
spectra, Simon5 warns that the dip (1160 cm ) in the broad reflection

peak cannot be considered as a gap between two partially overlapping

peaks, but rather a display of an individual, sharp absorption bond.
-1l+
The peak at approximately 1110 cm can be assigned to Si-O-Si vibration

from a single SiO4 tetrahedron with the symmetry point group Td. This

peak and peaks in other silicates from similar stretching vibrations

will be referred to in this work as S peaks. These results agree nicely

with assignment proposed for a vitreous Si04 tetrahedron in perfectly

symmetrical conditions.3
































Fig. 11-9. Polarized infrared reflection spectra of a
quartz single crystal on the prism face (1010).








































1300 1100 900
WAVENUMBER (cM-1)










-l
The peak at 1180 cm can be assigned to the Si-O-Si stretching

vibration between adjacent tetrahedra in the crystal structure. This

vibration from the lattice structure gives an indication of the long
63
range order of the single tetrahedral units in silicates. The basic

unit cell used for predicting the possible modes of stretching vibra-
-2
tion for lithia-disilicate is Si202 which contains two oxygen tetra-

hedra with one nonbridging oxygen each and both joined together by

a point. Figure II-10 shows the unit cell with the eight possible

stretching modes for the Si-0 bonds. The unit cells are shown oriented

with the primary axes with the characteristic polarization direction

indicated. For a more precise orientation of the unit cell with respect

to the crystal structure refer to Fig. 11-3. The frequency assignments

were made after analysis of the PIRRS data from the two mutually per-

pendicular planes (010) and (001), as shown in Fig. II-11 and 11-12,
-i
respectively. The one peak at 1115 cm (S) in all the spectra is

the SilOTSi stretching vibration that was predicted to have components in

all directions in the crystal. From the spectra along the X direction

in both of the planes, two peaks, 1230 cm and 1040 cm are obtained

in addition to the S peak. The predicted stretching vibrations are

the silicon-nonbridging oxygen along the X direction (Si-0 :X) and

the oxygen-silicon-oxygen along the X direction (Si-02:X). The silicon-

nonbridging oxygenvibrational frequency has been shown to be lower

than the S vibrational frequency.37,63 The peak assignments of 1040 cm-1
-1
to Si-0NB:X and 1230 cm to Si-02:X therefore follows. Along the

Z direction in Fig. II-11, the only peak besides the S is at 950 cm-1
Z direction in Fig. II-11, the only peak besides the S is at 950 cm




































-2
Fig. II-10. Oxygen tetrahedra for Si207 2 unit with one
nonbridging oxygen per tetrahedra illustrating
possible Si-O bond stretching vibrational modes.


















SI-O-SI
(1115 cM-1)


0-SI-0
(1230 M-1)


X POLARIZED


LITHIA-DISILICATE

N
NB


y


SI-O-SI NB
(INACTIVE) /



NB


X-
X


0-SI-0


0-SI-O


0 UPWARD MOVEMENT DOWNWARD MOVEMENT


































Fig. II-11. Polarized infrared reflection spectra of a
lithia-disilicate single crystal on the (010)
plane.






43














33L SINGLE CRYSTAL (010) PLANE
POLARIZED SPECTRA
----45" (X DIRECTION)
-315" (Z DIRECTION)


SI-02:Z

Sl-0--
SI X S...:X



-j










I
U-
LU|




LU
-- 2:XV








1300 1100 900

WAVENUMBER (CM-1)



































Fig. 11-12. Polarized infrared reflection spectra of a
lithia-disilicate single crystal on the (001)
plane.








































1300 1100 900
WAVENUMBER (cM-1)










and is assigned the oxygen-silicon-oxygen vibration along the Z direc-

tion (Si-02:Z). Finally, the Y direction has a peak at 995 cm-1 which

is assigned to the silicon-nonbridging oxygen vibration in that direc-

tion (Si-0NB:Y). The small peaks at 950 cm- and 1230 cm- are possibly

due to either imperfect crystal orientation in the infrared beam or

to the stretching vibrational mode polarization direction that is not

perfectly aligned with the crystal axis.
-2
The Si20-2 unit cell with two nonbridging oxygen per tetrahedra

was used to predict the possible modes of vibration for lithia-

metasilicate. Figure 11-13 shows the predicted vibrational modes and

polarization directions for infrared active Si-0 stretching modes.

The lithia-metasilicate samples contained long single crystals oriented

with their axis of growth (001 or Z axis) parallel. Since these crystals

are chain-link, the orientation of the X and Y directions was random.

The PIRRS data for lithia-metasilicate (50L) taken from planes parallel

to the Z axis and perpendicular to the Z axis are shown in Fig. II-14

and 11-15, respectively. The spectra taken along the Z axis in Fig.

11-14, shows two peaks at 870 cm- and 1000 cm The 870 cm peak

is due to the 0-Si-O stretching vibration along the tetrahedral chains

of the 50L structure (Si-02:Z). The 1000 cm-1 peak appears in all

spectra which would indicate that its polarization direction is not

precisely aligned with the principal axes. However, this peak is

assigned to the Si-(NB )2:X vibration because this vibration mode would

more likely be excited by radiation in the Z direction than the vibra-

tional modes in the Y direction. The two remaining peaks 1120 cm-




































-2
Fig. 11-13. Oxygen tetrahedra for Si207-2 unit with two
nonbridging oxygens per tetrahedra illustrating
possible Si-O bond stretching infrared vibrational
modes.






LITHIA-METASILICATE


SI-0-SI


SI-(ONB)2


SI-(ONB)2


NB
A


SIO2


SI-O-SI


NB


Y
Z


NB


NB


NB





NB
A


NB
N

































Fig. 11-14. Polarized infrared reflection spectra of lithia-
metasilicate oriented crystals on a plane parallel
to the crystal growth axis (001).






































1300 1100 900
WAVENUMBER (cM-1)


































Fig. 11-15. Polarized infrared reflection spectra of lithia-
metasilicate oriented crystals on a plane per-
pendicular to the crystal growth axis (001).






































1300 1100 900
WAVENUMBER (cM-1)










-l
and 965 cm- are due to the Si-O-Si stretching (S:Y), which is

polarized in this silicate, and the other silicon-nonbridging oxygen

vibration (Si-(ONB)2:Y), respectively. The S:Y peaks are polarized
-4
apparently due to the increase in symmetry of the SiO tetrahedral

positions in the metasilicate chain structure. The PIRRS data in Fig.

11-15 shows that perpendicular to the crystal growth direction, the

structure has a random orientation about the Z axis.



Conclusions

The polarized infrared reflection spectroscopic technique is very

useful in making vibrational mode assignments to spectral reflection

peaks from single crystal silicates. Table II-1 lists the spectral

peaks for quartz, lithia-disilicate and lithia-metasilicate. From

this summary, it can be seen that by increasing the number of non-

bridging oxygens per SiO4 tetrahedra (NBO) there are two different

types of frequency changes in the spectra. First with increasing NBO,

the frequency of the S type vibration and vibrations of the tetrahedral

lattice increase. Secondly, vibrations involving the Si-ONB bonds

and Si-02 bonds along tetrahedral chains decrease in frequency with

increasing NBO. This illustrates that one cannot predict a uniform

peak shift in the spectra due to changes in the silicate structure.














Table II-1. Vibrational Frequencies of Infrared Spectral
Peaks in the Range 1400-800 cm-1 for Quartz,
Lithia-Dislicate and Lithia-Metasilicate.


Lithia- Lithia-
Quartz Disilicate Metasilicate
Wavenumber Si02 33L 50L
(cm-1) (0-NBO) (1-NBO) (2-NBO) Peak Assignment


1230 X Si-O2:X Polarized

1180 X Si-02 (lattice)
+ +
1120 X Si-O-Si:Y Polarized

1115 X Si-O-Si

1110 X Si-O-Si

1040 X Si-0NB:X Polarized

1000 X Si-(ONB)2:X Polarized

995 X Si-ONB :Y Polarized

965 X Si-(ONB)2:Y Polarized

950 X Si-02:Z Polarized

870 X Si-02:Z Polarized


Note: NBO means nonbridging oxygen per tetrahedra.















CHAPTER III
INFRARED REFLECTION SPECTROSCOPY OF POLYCRYSTALLINE
LITHIA-DISILICATE GLASS-CERAMIC



Introduction

Glass-ceramics can be microstructurally considered as numerous

small crystals in a glass matrix. These polycrystalline materials

are typically prepared from preformed cast glass articles via a

specific nucleation-crystallization heat treatment. This process

normally produces a very fine crystal (1-.p) that is randomly oriented

in the matrix. For this study the stoichiometric Li20 2SiO2 (33L)

system was used, producing a glass-ceramic with the crystalline phase

and the glass phase having the same composition.85

Infrared reflection spectroscopy (IRRS) has been used by pre-

vious investigators for studying both compositional and structural

changes in glass systems.22'25'26'37'87'88 Furthermore, in another

study IRRS has been shown as a nondestructive analytical tool for

determining solid-phase reactions during the sintering of crystalline
24
silicates and devitrification of lithia-silica glasses. None of

these studies have given structural vibration assignments to the

infrared peaks of spectra from glass-ceramics.

The PIRRS technique discussed in Chapter II cannot be directly

applied to the study of glass-ceramic surfaces. For the PIRRS tech-

nique to be effective, the samples must be either single crystals











or oriented polycrystalline surfaces. However, the vibrational assign-

ment should be possible by comparing the IRRS spectra of the lithia-

disilicate glass-ceramic to the PIRRS spectra of the lithia-disilicate

single crystal.



Experimental

The glass composed of 33 mol % Li20 and 67 mol % SiO2 was pre-

pared from reagent grade Li2CO3* and 5 pm silica** by melting them

in a covered Pt crucible at 1350C for 24 h. Cylinders 6.0 cm long

and 2.5 cm in diameter were cast in a graphite mold and annealed at

230C for 4 h. The cast samples were nucleated for 24 h at 4750C

and then crystallized at 550C for 2 h, 8 h, and 24 h. The nuclea-

tion-crystallization treatments were performed in a nichrome-wound

tube furnace equipped with an aluminum block to provide a constant

temperature (1lC) over a 6-in. hot zone. The cylinders were wafered

using a diamond saw into disks 0.5 cm thick and 2.5 cm in diameter.

Subsequently, the disks were rough polished to 600 grit with SiC paper

and final polished to a 6 pm finish with diamond paste. The volume

fraction of crystals present in the partially crystalline samples

were measured by microscope using optical scanning techniques of

applied stereology discussed by Freiman.17 Both optical and scanning

electron microscopes were used with hydrofluoric acid etched samples

of each crystallization time. Infrared reflection spectra were





*Mallinckrodt, Inc., St. Louis, MO.
**Min-U-Sil, Pennsylvania Glass and Sand Corp., Pittsburgh, PA.










determined using a Perkin-Elmer Model 467 in the spectral region of

1400-800 cm- for all specimens after nucleation-crystallization.



Results and Discussion

The three crystallization times for the lithia-disilicate (33L)

glass-ceramic produced the following volume fraction of crystallinity:


2 h 20% crystalline (33L-20%)

8 h 60% crystalline (33L-60%)

24 h 90% crystalline (33L-90%)


The optical microscope revealed that no crystallization was apparent

in the lithia-disilicate glass (33L-0%).

The comparison of the PIRRS spectra of the 33L single crystal

with the IRRS spectra of the 33L-90% is shown in Fig. III-1. The

PIRRS spectra and peak assignments of the 33L single crystal are

discussed in Chapter II. The spectra from the 33L-90% glass-ceramic

has four peaks (1230 cm 1110 cm 1010 cm and 960 cm-1). These

data agree with published spectra of fine grained polycrystalline

lithia-disilicate.88 As can be seen from Fig. III-1, the spectral

peak assignment for 33L-90% glass ceramic is straightforward. The

1220 cm- peak is the 0Si-O stretching vibration while the 960 cm-1

peak is the other O-Si O stretching vibration. The sharp 1110 cm-1

is due to the Si-O-Si stretching vibration. The broad band centered

-1
about 1010 cm is generated by the overlapping of both Si-0NB

stretching vibrations.


































Fig. III-1. Infrared reflection spectra of 33L-90% crystal-
line glass-ceramic and polarized infrared
reflection spectra of 33L-single crystal from
all three axes--X, Y, and Z.









33L GLASS-CERAMIC
90% CRYSTALLINE


1220


33L SINGLE CRYSTAL
POLARIZED SPECTRA
- X DIRECTION
.**** Y DIRECTION
-- Z DIRECTION


1040
SI-ONB


1230
SI-Oo


1115


995
SI-ONB:Y
Si: 950
I IS -02:Z
-i SlO2:Z


1100 900
WAVENUMBER (cM-1)


1010


960


1300










The surface of the 33L-60% glass-ceramic is predominately crystal-

line. This is markedly shown in Fig. III-2 in the IRRS spectra since

33L-60% glass-ceramic is nearly identical to 33L-90% glass-ceramic.

The overall decrease in intensity is most likely due to the lower

volume fraction of the crystalline phase. Obviously the spectral

peak assignments for 33L-60% glass-ceramic are the same as those for

33L-90% glass-ceramic.

The IRRS spectra for 33L glass and 33L-20% glass-ceramic are

shown in Fig. III-3. The spectral peaks are broader and less defined

than those of the crystalline lithia-disilicate. This is mainly due

to the wide variation in the intertetrahedral angles (Si-O-Si) caused

by the amorphous nature of glasses.89 This type of infrared spectra

for the 33L glass has been well characterized by Sanders et al.37

The peak occurring at 1050 cm- is due to a symmetrical silicon-

oxygen stretch vibration (LS) in an alkali ion (lithium) environment.

The second peak occurring at 920 cm-1 is a result of the silicon-

nonbridging oxygen stretching vibration (NS).

With only 20% crystalline phase, the glass phase is the pre-

dominate structural feature on the surface of the 33L-20% glass-

ceramic. Thus, the 33L-20% spectra shows only a slight deviation

from the 33L-0% spectra.



Conclusions

The combination of infrared reflection spectroscopy (IRRS) and

polarized infrared reflection spectroscopy (PIRRS) provides an

































Fig. 111-2. Infrared reflection spectra of 33L-90% crystal-
line and 33L-60% crystalline glass-ceramics.





























LU
C__


LU
-j
LL.
LU



SI-O SI-02
LU

S




S102

f I I ,

1300 1100 900

WAVENUMBER (cM-1)


































Fig. 111-3. Infrared reflection spectra of 33L glass and
33L-20% crystalline glass-ceramic.
















33L GLASS AND GLASS-CERAMIC
- 0% CRYSTALLINE (GLASS)
..... 20% CRYSTALLINE












1050
LS
.... 920
I ltNS




I ** ** 1uI I


1300


1100


WAVENUMBER (cM-1)


900






65



effective tool for characterizing highly crystalline glass-ceramics,

provided that a single crystal of the crystalline phase can be

obtained for analysis by PIRRS, as shown by the assignment of vibra-

tion modes to the 33L-90% spectral peaks in Fig. I-i.

For glass-ceramics, IRRS is primarily sensitive to the dominant

phase on the surface of the sample studied. Infrared reflection

spectroscopy is not a good method for accurately determining the

volume fraction of crystallinity in glass-ceramics.















CHAPTER IV
AQUEOUS DURABILITY OF LITHIA-DISILICATE GLASS-CERAMICS



Introduction

The corrosion behavior of Li20.2SiO2 (33L) glass has been
20,37,52
studied extensively.203752 During the early stages of reaction

with a neutral aqueous environment, Li is selectively leached from

the glass surface via ion exchange with H (or H30 ) from the solu-

tion. In a closed system in which the ratio of glass surface area

to volume of solution (SA/V) is high, the pH of the solution will

increase with exposure time. If the pH is permitted to go above

approximately 9, a second mechanism of glass corrosion becomes

important in the corrosion behavior.90'91 This mechanism involves

dissolution of the glass network and is due to OH attack on the

silicon-oxygen bonds. The relative importance of the ion exchange

and network dissolution reactions is dependent on numerous factors

including solution pH, exposure temperature, time and (SA/V). The

corrosion mechanisms and kinetics of vitreous Li20*2SiO2 have been

studied by Sanders and Hench20 and Ethridge and Hench53 in a wide

range of environments. Additionally, the crystallization kinetics

for this glass have been well characterized by Hench et al.8 and

Freiman and Hench.51,92 It is well known that controlled nucleation

and crystallization of a glass can significantly improve the mechan-

ical properties of the resulting glass-ceramic compared to those











characteristic of the parent glass.55 During crystallization, the

microstructure of the glass is altered by the presence of both

crystals and phase boundaries. If the glass is only partially

crystallized, the microstructure may consist of isolated crystals

dispersed in a vitreous matrix.

The objective of this study is to establish the effects of con-

trolled crystallization on the chemical durability of the Li20"2Si02

glass-ceramic systems. The 33L composition has been chosen for two

reasons: 1) as already mentioned, the crystallization and the

corrosion behavior of the glass are well characterized, and 2) since

this is a stoichiometric composition, the crystalline phase has the

same composition as the glass phase. This stoichiometry permits the

evaluation of corrosion, independent of compositional variations that

may accompany crystallization of nonstoichiometric compositions.



Experimental

The Li20*2SiO2 (33L) glass and glass-ceramic specimens used in

this study were prepared using techniques described in Chapter III.

Briefly, the glass was melted in an electric muffle furnace at 13500C

for 24 h. Cylinders 2.5 cm in diameter and 6.0 cm long were cast

in a graphite mold and annealed at 3500C for 4 h. These glass

cylinders were nucleated at 4750C for 24 h and crystallized at 550C

for various times yielding 20%, 60% and 90% volume fraction (V )

crystallization. The percentage volume fraction crystallization was

determined by using optical scanning techniques of applied stereology.17










The cylinders were sliced with a diamond watering saw into 0.3 cm

thick disks, the surfaces of which were polished through 600 grit

with SiC paper prior to exposure.

Specimens of 33L glass and partially crystallized 33L glass were

exposed to an environment of either demineralized water, 0.1M NaOH,

or 0.1M HC1 maintained at 1000C* for up to 10 h. Duplicate specimens

were tested for each exposure condition to evaluate reproducibility.

The maximum difference in the solution data between any of the duplica-

tions was 46 ppm (265 25 ppm) of SiO2 and 11 ppm (135 5.5 ppm)

of Li+. The ratio of surface area of exposed material to solution
-1
volume (SA/V) was 2.0 cm Infrared reflection spectra** were

obtained for all specimens in the spectral region 1300-600 cm- both

prior to and after aqueous exposure. Scanning electron micrographs***

were also taken for selected samples. The corrosion solutions were

analyzed by easing pH, Li concentration,t and SiO2 concentration.ttt



Results and Discussion

Scanning electron micrographs for 33L glass containing four

volume fractions of crystallization (V = 0%, 20%, 60%, 90%) are

shown in Figs. IV-1 thru IV-5. Micrographs of uncorroded specimens




*Magni-Whirl Constant Temperature Bath, Blue M Electric Co.,
Blue Island, CT.
**Model 467, Perkin-Elmer Infrared Reflection Spectrometer,
Norwalk, CT.
***Model JSM-35C, JEOL Ltd., Tokyo, Japan.
tModel 801A, Orion Research, Inc., Cambridge, MA.
ttModel 603, Perkin-Elmer Spectrophotometer (atomic emission),
Norwalk, CT.
ttiModel DR-3803, Hach Chemical Co., Ames, IA.



































Fig. IV-1. Scanning electron micrographs of uncoated 33L
glass-ceramic with 0, 20, 60 and 90% crystal-
linity. Surface polished with 600 grit SiC
paper.












UNCORRODED 33L GLASS-CERAMICS














0% X-TAL 20 % X-TAL


4*9


90 % X-TAL


60O% X- TA L

































Fig. IV-2. Scanning electron micrographs of 33L glass after
2 h and 10 h exposure in acid, neutral, and base
aqueous solution at 1000C with SA/V = 2.0 cm-.














0% CRYSTALLINE


2 HOURS EXPOSURE
..,: W


~If


i


kii~


r'


*


10 HOURS EXPOSURE


ACID











NEUTRAL












BASE


































Fig. IV-3.


Scanning electron micrographs of 33L-20%
crystalline glass-ceramic after 2 h and 10 h
exposure in acid, neutral, and base aqueous
solutions at 100C with SA/V = 2.0 cm-1.
















2 HOURS EXPOSURE
;< : ,, :r -
I ': "':'


20( CRYSTALLINE


E 10 HOURS EXPOSURE


I


IrI


* -^ '" v

. :.y .' -, ^ *.. ..: .
, -s .



s.-- .. *^-: .-. -i


yl LJL .^--^^


)


NE 'I r-4










b^ E


~nL-
r.


cSiC


,~'


^w&


r
"
L-` ~






-If. l:' -I
76 tj. -



polishing scratches do not exhibit noticeable dimensional changes

during exposure to the acidic solution suggests that network dis-

solution is minimal in the low pH environment. In contrast, the

specimens exposed to the basic solutions exhibit broadened polishing

scratches and no surface cracks. Thus, the dominant mechanism of

corrosion is network dissolution. The small particles on the surface

are evidence of precipitation of some compound from the corrosion

solution. The specimens exposed to the neutral solution show indica-

tions of both ion exchange and network dissolution with the latter

being more apparent after long exposure times.

Three types of attack are observed on the glass-ceramic contain-

ing 20% crystallization (Fig. IV-3). In acidic solutions, only ion

exchange occurs as was the case with glass (V = 0%). In both

neutral and basic solutions network dissolution of the glassy phase,

and preferential attack of the boundary between the glassy and

crystalline phases are significant. The preferential phase boundary

attack may be due to stresses or compositional gradients at the inter-

face. Similar phase boundary attack has been reported by Baylor and
47
Brown for phase separated borosilicate glasses. Surface cracks

in the glassy phase of the specimen exposed to the neutral solution

indicates the presence of an ion exchanged layer even after 10 h of

exposure. These cracks do no occur in the crystalline phase. Possibly

the most important information provided by Fig. IV-3 is the relative

durabilities of the glassy and crystalline phases. The fully exposed

crystals shown in this figure demonstrate that the crystalline phase

is more resistant to network dissolution than the glassy phase.










(Fig. IV-1) show polished flat surfaces with some polishing scratches.

Figures IV-2 thru IV-5 illustrate representative surfaces for each

material after exposure to acidic, neutral and basic solutions.

Significant differences in the surface microstructural features as a

function of percent crystallization can be seen, particularly after

exposure to neutral and basic solutions. The crystalline phase is more

pronounced on the specimens exposed to the neutral and basic solutions

than on those exposed to the acidic solution. The similarity between

the surfaces exposed to either neutral or basic solutions is not

surprising because the pH of the neutral solution increases with

exposure time under static exposure conditions. The increase in

solution pH is due to exchange between Li+ ions from the solid and H

ions from the solution, resulting in an increase in OH ions in
21,23
solution.223 Thus, long exposure times to a static neutral solu-

tion favors network dissolution as does exposure to a basic solution

even for short times.

When V = 0% (Fig. IV-2), two types of corrosion are observed
v
over a wide range of pH values (pH = 1-13). In acidic solutions the

primary mode of corrosion is ion exchange. There are two features

in the micrographs in Fig. IV-2 that support this conclusion: 1)

surface cracks, and 2) polishing scratches. The cracks are related

to the ion exchange mechanism and usually appear after the specimen

has been removed from solution and permitted to dry, or after the

specimen has been subjected to a vacuum. The extent of ion exchange

required to produce surface cracks is not known but is probably

dependent on the composition of the glass. The fact that the


































Fig. IV-4.


Scanning electron micrographs of 33L-60% crystal-
line glass-ceramic after 2 h and 10 h exposure in
acid, neutral, and base aqueous solutions at 100C
with SA/V = 2.0 cm-1.
















60% CRYSTALLINE


2 HOURS EXPOSURE


10 HOURS EXPOSURE


ACID













NEUTRAL


'I

.I


BASE







































Fig. IV-5. Scanning electron micrographs of 33L-90% crystal-
line glass-ceramic after 2 h and 10 h exposure
in acid, neutral and base aqueous solutions at
100C with SA/V = 2.0 cm-1.


I













90% CRYSTALLINE


2 HOURS EXPOSURE

i"'G


10 HOURS EXPOSURE


ACID











NEUTRAL


BASE











As the crystalline phase increases from V = 20% to V = 60%
v V
and V = 90%, the surface morphology changes from that of isolated

crystals in a vitreous matrix to one dominated by the crystalline

phase (Figs. IV-4 and IV-5). This change is most easily seen on the

specimens exposed to the neutral and basic solutions. Network dis-

solution preferentially removes the glass and highlights the crystal-

line phase. The surface of the crystals appear to be degraded by

the high pH solutions as illustrated by the removal of material

between the layers in the crystal. This intracrystalline attack

possibly could be the result of the dissolution of a residual glassy

phase trapped between the crystal layers during crystal growth or the

preferential attack of the crystals along selected crystallographic

planes. Exposure to the acidic solution produces essentially no

network dissolution but does permit ion exchange as evidenced by the

surface cracks visible on the 10 h specimens in Figs. IV-4 and IV-5.

Infrared reflection spectra shown in Fig. IV-6 for 33L glass

exposed to a wide range of pH solutions are consistent with those

reported in previous work.37 60 In the uncorroded glass spectrum,
-1
the peak located at 1030 cm (LS) is due to symmetrical Si-O-Si

stretching vibrations in a network containing Li and the peak at
-1
930 cm (NS) is due to Si-nonbridging oxygen vibrations. The spec-

trum of vitreous silica is also included in this figure for the
-i
purpose of instrument calibration and comparison. The peak at 1100 cm

(S) for vitreous silica is due to Si-O-Si symmetrical stretching vibra-

tions in a pure SiO2 structure. During corrosion of 33L glass in



































Fig. IV-6.


Infrared reflection spectra of 33L glass before
and after acid, neutral, and base aqueous solu-
tion.























































1200 1000 800
WAVENUMBER (cm-1)











the acidic solution the LS peak shifts to higher wavenumbers and

increases in intensity (% reflection) progressively approaching the

S peak of vitreous silica as corrosion time increases. These altera-

tions in the spectra are caused by the leaching of Li+ (ion exchang-

ing with H+ or H30+) from the glass, resulting in the development

of a SiO2-rich film on the glass surface.22 Infrared spectra show

that the short time corrosion behavior of 33L glass in the static

neutral solution is similar to that in the acidic solution. That

is, initially there is a development of a SiO2-rich film on the sur-

face of the glass. However, the peak intensity decreases with long

exposure times (i.e., >4 h). This is usually indicative of surface
23
roughening due to network dissolution of the SiO2-rich film.2 As

discussed earlier, the pH of the static neutral solution increases

with exposure time, and the resulting high OH concentration enhances

network dissolution. The general shape of the infrared spectra for

33L glass exposed to the basic solution does not change. The spectral

intensity, however, continuously decreases with exposure time. These

spectral variations indicate that no Si02-rich film is developed

during exposure to the high pH solution. Surface roughening due to

network dissolution is responsible for the decline in intensity.

Network dissolution enlarges the polishing scratches and causes pit-

ting as shown in the micrographs in Fig. IV-2.

Figure IV-7 shows the infrared reflection spectra for 33L glass

with 20% crystallization exposed to the acidic, neutral and basic

solutions. These spectra are basically the same as those for the


































Fig. IV-7. Infrared reflection spectra of 33L-20% crystal-
line glass-ceramic before and after acid, neutral,
and base aqueous corrosion.




















































1200 1000 800
WAVENUMBER (cm-1)











glass exposed to similar environments. The major exceptions are the

spectra corresponding to the 10 h exposures in neutral and basic solu-

tions. The lower intensities observed on the 33L-20% V specimen

for these exposures indicate increased surface roughening. The

increased surface roughening is due to preferential network dissolu-

tion of the glassy phase exposing the crystals as shown in Fig. IV-3.

These data suggest that the glassy phase of 33L-20% V glass-ceramic
v
is primarily responsible for the observed surface corrosion.

Infrared reflection spectra for 33L glass with 60% V and 90%
v
V crystallization exposed to acidic, neutral and base solutions are
v
shown in Figs. IV-8 and IV-9, respectively. The similarity of the

spectra for these two materials both prior to and after corrosion

facilitates their joint discussion. These spectra show some definite

changes in surface structure and corrosion behavior compared to the

33L glass and 33L glass with 20% V crystallization. As discussed
V
earlier in Chapter III, the IRRS spectra obtained from uncorroded

33L-60% and 33L-90% consist essentially of vibrations from only the
-1
crystalline phase. The broad peak at 1000 cm- is due to the over-

lapping of the two Si-0NB vibrations found in the lithia-disilicate

single crystal. Intracrystalline attack, as seen in the SEM micro-

graphs, Figs. IV-4 and IV-5, results in the reduction of the Si-NB:Y
-i
peak at 995 cm This reduction is due to the preferential removal

of material along the micaceous layer (010) in the crystals. There-

fore, this exposes more X-Z planes for interaction with the infrared

radiation. The well defined peaks at 1210 1110 and 960 -1
radiation. The well defined peaks at 1210 cm 1110 cm and 960 cm




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs