close

Вход

Забыли?

вход по аккаунту

?

Microwave synthesis using multicomponent and multiphasic systems

код для вставкиСкачать
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. U M I
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter free, while others may be
from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted.
Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send U M I a complete
manuscript and there are missing pages, these will be noted.
Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy.
Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact U M I directly to
order.
UMI
A Bell & Howell Information Company
300 North Zeeb Road, Ann Arbor M I 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Pennsylvania State University
The Graduate School
M ICRO W AVE SYNTHESIS USING M ULTICOM PONENT AND
MULTIPHASIC SYSTEMS
A Thesis in
Materials
by
M ilton D. Mathis
Submitted in Partial Fulfillment
o f the Requirements
for the Degree o f
Doctor o f Philosophy
May 1997
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 9732325
Copyright 1997 by
Mathis, Milton Douglas
All rights reserved.
UMI Microform 9732325
Copyright 1997, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, M I 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
We approve the thesis o f M ilton D. Mathis.
Date o f Signature
J o ** • / % / ? ?
D inesnK.
nesh*k. Agrawal
Associate Professor o f Materials
Thesis Co-Advisor
fZ
RustunyRoy
E v a n /u g h Professor o f the Solid State
Thesis Co-Advisor
Chair o f Committee
^
in?
W illiam B. White
Professor o f Geochemistry
Ik tz L J L u j y
Ross Plovnick
Senior Research Specialist
3M Company
Special Member
Jonn Jewett Henry
Professor o f Mechanical Engineering
f t • 7 W '/
y- ''M
r 7
Robert N. Pangbom
Professor o f Engineering Mechanics
In Charge o f Graduate Program in Materials
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
The research described in this thesis is focused on the microwave synthesis o f the
precursors o f technologically important ceramic materials.
The materials selected in
this work include aluminum titanate (A liT iO s), a low thermal expansion material:
barium titanate (BaTiC^), the most important electroceramic material: lead zirconate
titanate (P Z T ) and lead zinc niobate (P Z N ). contemporary electroceramic materials:
barium magnesium tantalate (B M T ), a low loss microwave substrate: silicon aluminum
oxynitride (S iA IO N ), an oxidation-resistant refractory material: zircon (Z rS iO j). a
traditional refractor)’ ceramic; and titanium diboride (TiB?). a refractory non-oxide
material.
A ll materials were synthesized, with the exception o f S iA lO N and zircon, starting
with non-stoichiometric phases using reduced oxide precursors. It was found that the
use o f partially reduced powders in microwave synthesis greatly enhanced the kinetics
o f the synthesis process.
Al^TiOs was microwave synthesized at 900°C in less than 15 minutes total time
including a soaking time o f only 1 minute at initial temperature using the "nonstoichiometric” starting materials.
Microwave synthesis o f the stoichiometric phase
required at least a 15 minute soak at 1300°C.
Conventionally synthesized Al;TiO?
required at least 1 hour soak time at 1500°C for the completion o f the reaction.
The
microstructure o f the partially sintered "non-stoichiometrically processed” material was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
nearly identical to the conventionally processed material.
Both samples microcracked
during cooling, which is an intrinsic characteristic o f AI^TiCh ceramics.
Microwave
synthesized
BaTiO j by non-stoichiometric routes nucleated
hexagonal phase at 300 °C with no soak time.
the
The reaction was completed by 700°C
with the formation o f the tetragonal BaTiC^ phase. The total time necessary for BaTiCb
synthesis via this route was less than 12 minutes.
occurs above
1300°C.
BaTiCb
Conventional synthesis o f B aTiO )
samples synthesized at
1000°C
by the "non-
stoichiometric'’ method show microstructural evidence o f some melting.
B a(M go 33Tao 67)03 (B M T ) was microwave synthesized by the "non-stoichiometric"
route using powder precursors o f B a fO H jix H jO , M gO and Ta^O?.,.
This reaction,
unlike the other "non-stoichiometric" syntheses, required the use o f a secondary coupler
in the microwave cavity because after the formation o f B M T as the major phase during
reaction, the sample ceased to couple with the microwaves. This is due to the fact that
phase pure stoichiometric B M T is itself a microwave transparent material.
The
microstructure o f a B M T sample heated to 1200CC shows partial sintering o f the
material.
S iA lO N was microwave synthesized at as low as 1500°C in 15 minutes o f soak
time using S ijN 4 and A I 2O 3 powders. In the conventional synthesis temperatures in
excess o f 1700°C are required to form S iA lO N .
Microwave synthesized S iA lO N also
showed submicron size particles whereas conventionally synthesized material consisted
o f long (ave. grain size >50pm ) acicular grains.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Microwave synthesis o f zircon (Z rS i0 4) was not successful.
It was found that
zircon synthesis could occur in a microwave field, but only using Z rO : board as a
secondary coupler.
the reactants.
However, the ZrC>2 board was found to preferentially couple over
Hence, zircon synthesis in the microwave was attributed to the
conventional heating from the secondary couplers used.
PZT was microwave synthesized by the “non-stoichiometric'' method using
PbCCb. Z rO ; and T iO :., powders.
X R D analysis showed that the reaction was near
completion at 600°C with only trace amounts o f PbO and ZrO ; detectable.
By 900°C
only very small amounts o f ZrO : remained. The cubic perovskite phase was formed at
600°C first, followed by the nucleation o f the tetragonal phase.
mixture o f the cubic and tetragonal phases was evident.
By 900°C a 50-50
The total time required to
achieve synthesis at 600°C in the microwave field was less than 8 minutes.
In summary, this study has demonstrated a new and very interesting aspect o f the
microwave processing o f materials.
W hile
most studies involving
microwave
interaction with ceramics have focused primarily on making dense ceramic bodies, this
study has focused on the synthesis o f powders o f various ceramic materials o f great
interest. Not only are the time and energy savings enormous, but we have demonstrated
a new strategy, o f reducing the variable valenced atoms, to achieve excellent microwave
coupling and very rapid heating. N ew reaction pathways have been found, including
possible metastable melting.
Many significant electroceramics, and other high use
materials have been synthesized in this study with savings in energy and time.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VI
T A B L E OF C O N T E N T S
LIS T O F F IG U R E S ......................................................................................................................xiii
LIS T O F T A B L E S .....................................................................................................................xviii
A C K N O W L E D G M E N T S .......................................................................................................... xx
C H A P T E R I : IN T R O D U C T IO N ................................................................................................. I
1.1 General....................................................................................................................................... 1
1.1.2 Statement o f Problem ,
1.1.3 Research Objectives...
C H A P T E R 2: L IT E R A T U R E R E V IE W ................................................................................... 6
2.1 Historical perspective on microwave usage in materials research................................. 6
2.2 Microwave-Material Interactions T h eo ry ...........................................................................10
2.3 Microwave Processing............................................................................................................16
2.3.1 Review o f microwave applicators......................................................16
2.3.2 Review o f microwave sintering and processing..............................22
2.3.3 Review o f microwave synthesis techniques.................................... 32
2.3.4 Review o f novel microwave applications and patents................... 36
C H A P T E R 3: E X P E R IM E N T A L .............................................................................................. 43
3.1
Microwave equipm ent.........................................................................................................43
3.1.1 Panasonic 900 W multimode microwave furnace.......................... 43
3 . 1. 1.2
Temperature measurement....................................................43
3.1.2 F iv e -K W multimode u n it.................................................................... 43
3.1.3 M M T M odel 10-1300........................................................................... 44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 .1.3.1 Microwave supporting materials............................................. 48
A ) Insulation.......................................................................................... 48
3.1.3.2 Temperature measurement........................................................50
3.2 Chemical preparations............................................................................................................51
3.3 Material characterization techniques................................................................................... 51
3.3.1 X R D (Powder, high-temperature and quantitative).........................51
3.3.2 Dielectric measurements........................................................................52
3.3.3 TG-GC-1R measurements..................................................................... 53
3.3.4 Electrical conductivity measurements................................................53
3.3.5 Microstructure examination.................................................................. 55
C H A P T E R 4: M IC R O W A V E S Y N TH E SIS OF A L U M IN U M T IT A N A T E ................... 56
4.1 Introduction............................................................................................................................... 56
4.1.1 General...................................................................................................... 56
4.1.2 Applications o f aluminum titanate..................................................... 59
4.1.3 Synthesis techniques.............................................................................. 60
4.1.3.1
Conventional synthesis o f aluminum titanate.................... 60
4.1.4 Microwave synthesis o f aluminum titanate...................................... 62
4.2 Experimental procedures.......................................................................................................63
4.2.1 Alumina-titania (anatase) system........................................................63
4.2.2 High temperature X R D analysis o f the anatase-alumina reaction63
4.2.3 Conversion o f anatase to defective ru tile .......................................... 64
4.2.4 Alumina-defect titania (rutile) system................................................64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VIII
4.2.5 Conventional synthesis o f aluminum titanate.................................. 64
4.2.6 Microwave synthesis o f A ljTiO ? in air and nitrogen......................65
4.2.7 Microwave synthesis o f aluminum titanate from reduced ru tile . 66
4.3
Results and Discussion....................................................................................................... 66
4.3.1 High temperature X R D analysis o f the anatase and alumina
reaction.............................................................................................................. 66
4.3.2 Conventional and microwave A^TiO?
synthesis in a ir........... 68
4.3.3 Conventional and microwave A I 2T 1O 5 synthesis in nitrogen...... 70
4.3.4 Aluminum titanate synthesis from defect rutile in nitrogen......... 75
4.3.5 Microstructural comparison o f defective and conventional
A lJ iO s ...........................................................................................................
86
4.3.6 Reoxidation o f defective aluminum titanate.................................... 86
4.4
Summary.................................................................................................. 00
C H A P T E R 5: M IC R O W A V E S Y N TH E S IS OF B A R IU M T IT A N A T E ......................... 91
5.1 Introduction..............................................................................................................................91
5.1.1 General.....................................................................................................91
5.1.3 Applications............................................................................................96
5.1.4 Synthesis................................................................................................. 98
5.2 Experimental Procedures...................................................................................................... 99
5.2.1 B aC O j-T iO ; system.............................................................................. 99
5.2.2 BaC 0 3 -T i 0 2 .x system......................................................................... 100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.3 Conventional synthesis o f Barium Titanate.................................... 100
5.2.3.1 Using BaC 0 3 -T i 0 :.x................................................................ 100
5.2.4 Microwave synthesis o f barium titanate..........................................101
5.2.4.1 Using B aC O j-T iO ; powders..................................................101
5.2.4.2 Using BaC 0 3 -T iO ;.x powders............................................... 102
5.3 Resultsand discussion.......................................................................................................... 103
5.3.1 T G A -G C -IR analysis o f the conventional BaC 0 3 -T iO : reactionl03
5.3.2 Conventional BaTiCb synthesis by solid state reaction using
T iO :.x................................................................................................................103
5.3.3 Stoichiometric microwave synthesis............................................... 107
5.3.4 Microwave synthesis o f BaTiCb from non-stoichiometric T iO ; 108
5.3.5 Heating effects in non-stoichiometric synthesis.......................... 110
5.3.6 Microstructural examination o f BaTi0 3 specimens................... 113
5.4 Sum m ary................................................................................................................................. 115
C H A P T E R 6 : M IC R O W A V E SY N TH E SIS O F B A R IU M M A G N E S IU M
T A N T A L A T E ( B M T ) ............................................................................................................... 116
6.1 Introduction...........................................................................................................................116
6.1.1 General................................................................................................. 116
6.1.2 Applications.........................................................................................116
6.1.3 Synthesis techniques............................................................................117
6.2 Experimental Procedures.................................................................................................... 118
6.2.1
Synthesis o f stoichiometric B M T from B a(OH):. M gO and Ta^O?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X
118
6.2.2 Reduction o f tantalum pentoxide.....................................................119
6.2.3 Synthesis o f B M T using non-stoichiometricT a :0 < ..................... 119
A. BaC 0 3 -M g 0 -Ta 205 .Ksystem........................................................ 119
B. B a(O H )2 x H 2 0 -M g 0 -Ta 205 -x system.......................................... 120
C. Ba 0 -M g 0 -Ta 205 system.................................................................120
6.2.4 M icrowave synthesis o f B M T from stoichiometric Ta:0< routes
120
6.2.5 M icrowave synthesis o f B M T from non-stoichiometric T a; 0 5 . 121
6.3 Results and Discussion..................................................................................................... 122
6.3.1 Synthesis o f reduced Ta 2 0 < powders................................................122
6.3.2 Microwave synthesis o f non-stoichiometric B M T using BaCO. 122
6.3.3 Microwave synthesis o f non-stoichiometric B M T using a
Ba(O H )2 route..................................................................................................126
6.3.4 Microwave synthesis o f B M T using BaO and Ta;0<.x................131
6.4 Conclusion.............................................................................................................................. 132
C H A P T E R 7: M IC R O W A V E S Y N T H E S IS OF S IA L O N .................................................134
7.1 Introduction............................................................................................................................. 134
7.1.1 General.................................................................................................... 134
7.1.2 Applications...........................................................................................136
7.1.3 Synthesis.................................................................................................137
7.2 Experimental...........................................................................................................................139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7.2.1 Clay-carbon-nitrogen system............................................................. 139
7.2.2 A lO O H -S iO i-S iC -C -N : (BSSCN) system......................................139
7.2.3 Si3N 4-A h O j system.............................................................................139
7.3 Results and Discussion........................................................................................................ 140
7.3.1 Synthesis using Clay-Carbon-Nitrogen........................................... 140
7.3.2 Synthesis using Boehmite-Silica-SiC-Carbon-Nitrogen (B S S C N )
........................................................................................................................... 141
7.3.3 Synthesis using silicon nitride and alu m in a................................... 144
7.3.4 Microstructure analysis o f S ijN ^ A liO ] route................................146
7.4 Sum m ary.................................................................................................................................150
C H A P T E R 8 : M IC R O W A V E S Y N T H E S IS OF P Z T ........................................................ 152
8.1 Introduction............................................................................................................................ 152
8.1.1 General....................................................................................................152
8.1.2 Applications.......................................................................................... 154
8.1.3 Synthesis Techniques......................................................................... 155
8.2 Experimental Procedure.......................................................................................................157
8.2.1 Microwave synthesis o f PZT using reduced T iO ; .........................157
8.3 Results and discussion......................................................................................................... 158
8.3.1 Microwave synthesis o f P Z T .............................................................158
8.3.2 Microstructural analysis o f P Z T ....................................................... 159
8.4 Sum m ary.................................................................................................................................160
C H A P T E R 9: O T H E R S Y S T E M S IN V E S T IG A T E D .........................................................164
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
xii
9.1 Introduction............................................................................................................................ 164
9.2 Zircon (ZrS iO ^ )..................................................................................................................... 164
9.2.1 Experimental Procedure.................................................................... 165
9.2.1.1 Solid state synthesis.................................................................165
9.2.1.2 Sol-gel synthesis.......................................................................165
9.2.2 Results and Discussion.......................................................................166
9.2.2.1 High temperature X R D analysis o f the ZrO :-SiO : reaction
166
9.2.2.2 Microwave synthesis o f ZrSiOa............................................ 167
9.3 Titanium diboride ( T iB ; ) .................................................................................................... 168
9.3.1 Experimental procedure..................................................................... 169
9.3.2 Results and discussion........................................................................169
C H A P T E R 10: G E N E R A L C O N C L U S IO N S ...................................................................... 171
10.1 Suggestions for Future W o rk ........................................................................................... 175
10.2 Concluding Statement....................................................................................................... 176
R E F E R E N C E S ............................................................................................................................178
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XIII
LIST OF FIGURES
Figure 2.1. The Electromagnetic Spectrum and Response Mechanisms o f Molecules ...7
Figure 2.2. Response Mechanism o f Water to a Microwave Field when (a) the Field is
Applied and (b) the Field is R elaxed.........................................................................................12
Figure 2.3. Interaction o f Microwaves with Materials..........................................................15
Figure 2.4. Single Mode TE|o„ Microwave A pplicator....................................................... 17
Figure 2.5. Single Mode TEoi i Microwave A pplicator.........................................................18
Figure 2.6. Single Mode TMoio Microwave A pplicator.......................................................20
Figure 2.7. Multimode Microwave Applicator......................................................................2 1
Figure 2.8. Mode Set-up in a Multimode Applicator............................................................ 23
Figure 2.9. Comparison o f Microwave and Conventional Sintering o f 1la p ................... 24
Figure 2.10. M icrowave Sintering of Alumina at 28 G H z at Various Temperatures ....26
Figure 2.11. The Relationship between Frequency and Loss Tangent at Room
Temperature................................................................................................................................... 27
Figure 2.12. Comparison o f Arrhenius Plots for Microwave and Conventional Sintering
o f Alum ina at 28 G H z .................................................................................................................. 28
Figure 2.13. Arrhenius Plot o f the Microwave-Driven Diffusion o f O 18 into Single
Crystal Sapphire.............................................................................................................................30
Figure 2.14. Calculated and Measured Interdiffusivities for the Alumina-Chromia
System..............................................................................................................................................31
Figure 2.15. Comparison o f the Curve fitting o f the Alum ina-Zinc Oxide Reaction to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
x iv
the Valensi-Carter Model in (a) M icrowave and (b) Conventional Applicators............. 35
Figure 2.16. The Microwave Dilatometer............................................................................. 39
Figure 2.17. The Combination Microwave-Conventional Furnace................................. 42
Figure 3.1. The Microwave Materials Technologies Model 10-1300 Microwave Unit 46
Figure 3.2. Schematic o f the Atmosphere Chamber for the Model 10-1300 U n it......... 47
Figure 3.3. Typical Set-up for Processing in a Microwave Furnace.................................48
Figure 3.4. Schematic o f the Furnace and the Hewlett-Packard Network Analyzer used
for Obtaining Dielectric Measurements................................................................................... 54
Figure 4.1. Binary Phase Diagram o f the System A I 2O 3- T O ; ........................................... 57
Figure 4.2. The Crystal Structure o f p *A l;T i 0 5 ....................................................................58
Figure 4.3. X R D pattern o f P -A i^T iO f.................................................................................... 61
Figure 4.4. X R D Patterns for A I;O i and T iO ; heated at various temperatures............. 67
Figure 4.5. Dielectric Measurements o f the Reaction between A F O ; and T iO ; as
Temperature Increases in A ir ......................................................................................................73
Figure 4.6. Dielectric Measurements o f the Reaction between A I 2O 3 and T iO ; as
Temperature Increases in N itrogen........................................................................................... 74
Figure 4.7. Dielectric Measurements o f the Reaction between A l ;0 3 and T iO ;.x as
Temperature Increases in N itrogen........................................................................................... 77
Figure 4.8. The Heating Rate Plot o f the Reaction between A l ;0 3 and T iO ;.x
heated in Nitrogen.........................................................................................................................79
Figure 4.9. The Heating Rate Plot o f the Reaction between A I 1O 3 and T iO ; heated in
A ir ...................................................................................................................................................80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XV
Figure 4.10. The Heating Rate Plot o f the Reaction between A U O 3 and T iO ;
heated in N itrogen..........................................................................................................................81
Figure 4.11. The Heating Rate Plot o f T iO ;., heated in Nitrogen........................................ 82
Figure 4.12 Dielectric Measurements o f T iO ;., as Temperature Increases in Nitrogen .84
Figure 4.13. Dielectric Measurements o f T iO ;., as Temperature Increases in Forming
G a s .................................................................................................................................................... 85
Figure 4.14. S E M Micrograph o f the Microwave Reaction between A I 1O 3 and
T iO ;., heated at 1250°C for 15 minutes.................................................................................... 87
Figure 4.15. S E M Micrograph o f the Conventional Reaction between A l;0 ', and
T iO ;., heated in Nitrogen............................................................................................................. 88
Figure 4.16. Dielectric Measurements o f the Microwave Reaction between A l;O i
and T iO ;., in Nitrogen, then Under A ir Purge......................................................................... 89
Figure 5.1. The Binary- Phase Diagram o f the B aO -TiO ; System.......................................92
Figure 5.2. Phase Transitions o f BaTi0 3 .................................................................................93
Figure 5.3. Electrical Polarization o f T i 0 6 Octahedra.......................................................... 95
Figure 5.4. Schematic o f a Multi-layer Capacitor..................................................................97
Figure 5.5. T G A Thermogram o f the Reaction between BaC 0 3 and T iO ; .....................104
Figure 5.6. The Heating Rate Plot o f the Reaction between BaC 03 and T iO ;., heated in
N itro g e n ........................................................................................................................................ I l l
Figure 5.7. The Heating Rate Plot o f the Reaction between BaC 0 3 and T iO ; heated in
N itro g e n ........................................................................................................................................ i 12
Figure 5.8. S E M Micrograph o f the Reaction M ixture from BaC 0 3 and T iO ;., heated at
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XVI
1000°C in Nitrogen for 15 M inutes....................................................................................... 114
Figure 6 .1. X R D Patterns for (a) T ajO j and (b) T a jO j.x .................................................. 123
Figure 6.2. The Heating Rate Plot o f the Reaction between BaCCb, M gO and T a:0«.x
heated in Nitrogen......................................................................................................................125
Figure 6.3. The Heating Rate Plot o f the Reaction between B a (O H ):. M gO and Ta: 0 5 .x
at 766°C in Nitrogen...................................................................................................................127
Figure 6.4. The Heating Rate Plot o f the Reaction between B a (O H ):. M gO and T a; 0 <
heated in A ir ................................................................................................................................. 129
Figure 6.5. The Heating Rate Plot o f the Reaction between B a (O H ):. M gO and TaiO<.x
heated in Nitrogen....................................................................................................................... 130
Figure 7.1. Phase Diagram for the S i-A l-O -N System........................................................135
Figure 7.2. SEM Micrograph o f the Products from the Microwave Reaction between
Si}N 4 and A I 1O 3 at 1500°C for 15 minutes.............................................................................147
Figure 7.3. SEM Micrograph o f Products from the Conventional Reaction between
Si3N 4 and A l :0 3 at 1648°C for 48 minutes............................................................................148
Figure 7.4. SEM Micrograph o f Products from the Conventional Reaction between
Si3N 4 and A I 2O 3 at 1700°C for 1 H o u r...................................................................................149
Figure 8.1. The Binary Phase Diagram for the PbTiCb-PbZrCb System (reference [s4])
................................................................................................................................................. 153
Figure 8.2. SE M Micrographs o f PbCCb, Z rO : and T iO :.x Microwave Heated to 800°C
in Nitrogen Showing the Beginnings o f Sintering............................................................... 161
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
xv i i
Figure 8.3. S E M Micrographs o f PbCOj, ZrO : and T iO :., Microwave Heated to 800°C
in Nitrogen Showing Grain G iow th .........................................................................................162
Figure 8.4. S E M Micrographs o f PbCOj, Z rO : and T iO ;., Microwave Heated to 800°C
in N itro gen ................................................................................................................................... 163
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
x v ii i
L IS T O F TA BLES
Table 2.1 The microwave frequencies allowed for industrial, scientific and medical use8
Table 2.2 The evolution o f microwave technology................................................................. 9
Table 4.1. X R D Analysis o f the Microwave Synthesis o f A liT iO s in A ir....................... 69
Table 4.2. X R D Analysis o f the Conventional Synthesis o f A l^TiO f in A ir ...................69
Table 4.3 X R D Analysis o f the M icrowave Synthesis o f AhTiO s in Nitrogen.............. 70
Table 4.4 X R D Analysis o f the Conventional Synthesis o f A ^T iO s in N itro gen
70
Table 4.5. X R D Analysis o f the Microwave Synthesis o f AliTiO ? using A I 2O 3 and
Defect T iO : in Nitrogen............................................................................................................... 75
Table 5.1. X R D intensities from the conventional BaCC>3-T iO :.x synthesis.................. 106
Table 5.2. X R D intensities from the microwave BaC 0 3 - T i(> reaction..........................108
Table 5.3. X R D intensities from the microwave BaC 0 3 -T i 0 :.x reaction........................109
Table 6.1. X R D peak intensities from the B aC C b-M gO -TaiC k, synthesis.................... 124
Table 6.2. X R D peak intensities from the Ba( 0 H ):-M g 0 -Ta: 0 5 .x reaction (without use
o f S iC ) ............................................................................................................................................126
Table 6.3. X R D peak intensities from the B aO -M gO -TaaC k, reaction..........................132
Table 7.1. X R D Results o f the M icrowave Clay-Carbon-Nitrogen Reaction................140
Table 7.2. X R D Results o f the Conventional Clay-Carbon-Nitrogen Reaction........... 141
Table 7.3. X R D Results o f the Microwave Reaction o f BSSCN System...................... 142
Table 7.4. X R D Results o f the Conventional Reaction o f B S S C N .................................143
Table 7.5. X R D Results o f the M icrowave Reaction between S ijN j and A I 1O 3 ..........145
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
xix
Table 7.6. X R D Results o f the Conventional Reaction between Si 3N.» and A I 1O 3
145
Table 8 .1. X R D peak intensities from the PbC O j-ZrO : -T iO : synthesis........................ 158
Table 9.1. High Temperature X R D Analysis o f Zircon synthesis....................................166
Table 9.2. X R D Peak Intensities o f Zircon Synthesized by Microwave Furnace Using
Z rO : Board (N D -no reliable data available).......................................................................... 168
Table 9.3 X R D peak intensities from the synthesis o fT iB : from T iO :.x and B4C
169
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
XX
ACKNOW LEDGM ENTS
This author would like to express gratitude and admiration for his co-advisors Dr.
Rustum Roy, Dr. Dinesh K. Agrawal and Dr. Ross H. Plovnick for their understanding,
guidance and support in research, thesis writing and personal matters. The author would
also like to thank Dr. W illiam White, Dr. John J. Henry and Yvonne Mathis for their
helpful comments and suggestions in matters concerning the revision o f this thesis. The
author would like to thank Brian Lynch and Myles Brostrom o f 3 M Analytical
Laboratories, as well as Dr. Chris Goodbrake o f 3M Ceramic Technology Center for
their support in x-ray diffraction and scanning electron microscopy characterization. The
author would like to thank Dr. Ron M . Hutcheon for his assistance in dielectric
characterization. The author would also like to thank Tom Forester and Kathy Humphal
o f 3M Ceramic Technology Center, Dr. Jipeng Cheng and Dr. Yi Fang o f the Materials
Dept, at Penn State for their assistance in experimental work. The author would also
like to thank the other members o f the 3 M Ceramic Technology Center for their advice
and for surrendering equipment time and resources for the continuance o f this work.
The author would like to thank all o f his friends and confidants, particularly Dr.
Kamau wa Gachigi and Dr. Judith LaRosa-Thompson, for all o f their support and aid
during this work.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chapter 1
IN T R O D U C T IO N
1.1 General
M icrowave synthesis and sintering o f materials is a relatively new and potentially
important way o f materials processing compared to heating in electric resistance
furnaces and other methods o f conventional processing. In fact, microwave processing
has already shown some promise in the synthesis and sintering o f some specialized
ceramics systems.
The main advantages o f microwave sintering technology over
conventional methods have been said to be: 1) the material is heated internally and
uniformly rather than from heat conduction: 2 ) very high heating rates are achieved
hence grain growth can be controlled resulting in better mechanical properties: 3)
microwave processing is highly efficient and environmentally benign: 4) due to last heat
rates and a novel method o f heat generation, microwave processing may provide new
fast synthesis techniques to form better and less expensive materials.
M icrow ave heating o f ceramics goes back to the post-World War II era when old
radar equipment was used for drying large whitewares.
Early work on the use o f
microwave sintering was concentrated on dark ceramics, especially magnetic materials,
as discussed in the review by Sutton1.
Extraordinarily rapid heating and high
temperatures were reported by Haas et. al. in urania gels. The first report on the same
effects, including thermal runaway, in white ceramics (specifically A U O 3 and silica)
were by Yang, Komameni and Roy.
Subsequent to the initiation o f this study,
beginning in about 1990, a wide range o f symposia have brought together an enormous
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
variety o f papers on the practices and theory o f microwave sintering o f white ceramics
and other materials.
These are reported in several volumes o f symposia proceedings
starting in 1991.
The first serious interest in microwave research formally appears in the 1960's.
The International Microwave Power Institute (IM P I) was the first society developed for
the purpose o f furthering knowledge in the various aspects o f microwave science and
technology, from food processing and cooking to the processing o f materials for
industrial, scientific and communication applications.
After the creation o f IM P I.
several other research societies began providing useful forums and organizing symposia
for microwave and electromagnetic research and processing.
The interest initiated by
symposia sponsored by IM P I. the Materials Research Society (M R S ), the American
Ceramic Society (ACerS). the American Chemical Society (AC S), as well as other
societies and industries nationally and internationally, has culminated in planning for the
first international microwave symposium and exposition to be held in January. 1997 in
Orlando. Florida. This symposium w ill be the first o f its kind providing individuals o f
various disciplines and backgrounds the opportunity to meet, present views and discuss
various aspects o f microwave technology.
1.1.2 Statement o f Problem
Historically, most o f the work done in the area o f microwave processing deals with
the sintering, drying and joining o f materials.
Though these issues are o f vital
importance to scientific and industrial researchers, some other areas have not yet
attracted as much attention. One such area is the microwave synthesis o f materials.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
M icrow ave synthesis, as microwave sintering has demonstrated, may provide solutions
for various laboratory and industrial problems in materials synthesis by possibly
enhancing the kinetics and changing the reaction routes o f synthetic reactions.
Currently, there is very little work reported in the area o f microwave powder
synthesis as opposed to microwave sintering and processing. But microwave synthesis
may offer many advantages over the conventional methods in terms o f cost, energy and
time savings, and the quality o f the product.
M any researchers have reported such
savings for materials joining and sintering, which could be essential for the ceramics
industry. However, if one could synthesize and then sinter materials using microwaves,
then these savings could be further augmented for the benefit o f the entire ceramic
industry.
One o f the necessities for successful microwave synthesis or processing is the
ability o f one or more o f the phases to couple with the applied field, preferably at room
temperature.
This would provide the basis for microwave-material interaction that
would result in heat generation followed by material synthesis, even i f there are other
phases present in the material that do not couple effectively. I f one phase couples with
the microwave field, that phase could heat, and react with, the other phases present in
the system by thermal conduction until the other phases also begin to couple with the
field. Finally, since the reaction routes and mechanisms are often difficult in that they
are dependent on reaction conditions, it may well be possible to synthesize new phases.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
1.1.3 Research Objectives
The systems selected for microwave synthesis in this work are commercially and
technologically very important: aluminum titanate ( A l : T i0 5), barium titanate (B a T iO i),
lead zirconate titanate (PZT), barium magnesium tantalatc (B M T ), silicon aluminooxynitride (S iA lO N ), titanium diboride (T iB ;) and zircon (Z rS i0 4).
systems has wide applications in industry.
Each o f these
However, with the exceptions o f A l’ TiO?.
B aTiO j and PZT. there are no published data on the microwave synthesis o f these
systems.
Even though AliTiO? and B aTiO j have been microwave synthesized by other
researchers, various important aspects o f the synthesis process have been ignored.
For
the first time, we have attempted to solid state synthesize various materials using metal
oxide phases that have been reduced to some degree, producing oxygen vacancies and
non-stoichiometric compounds.
Pure stoichiometric metal oxide precursors, such as
TaiCK, Nb^O? and T iO i. do not couple with microwave energy very efficiently unless
they are heated to high temperatures (>1000°C ). However, by partially reducing these
phases to oxygen defective states, such as TajC X *, NbiOs.* and T iO ;.x. it is possible to
increase their efficiency to absorb microwave energy at low temperatures dramatically
due to a large increase in their electron conductivities by reduction o f the metallic ions.
Hence, the main scope and objectives o f this work are to: (1) Synthesize
technologically important ceramic phases by microwave processing; ( 2 ) develop a novel
method to enhance microwave efficiency using reduced oxide precursors, and (3)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
accomplish a comparative study between conventionally and microwave processed
powders.
This work w ill also extensively study the reaction pathways and mechanisms in
each material synthesis from “non-stoichiometric” routes and compare them with
reaction pathways from stoichiometric microwave synthesis and conventional methods.
X-ray diffractometry (X R D ) w ill be applied to study the reaction paths, and phase
composition during various stages o f the synthesis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
Chapter 2
L IT E R A T U R E R E V IE W
2.1 Historical perspective on microwave usage in materials research
Microwaves are electromagnetic waves having wavelengths between 1m to 1mm.
which correspond to the frequencies from 0.3 to 300 G H z 1'2.
The location o f this
radiation in the electromagnetic spectrum is shown in Figure 2 .13. O f the frequencies
allocated by the Federal Communications Commission (FC C ), the most widely used
frequencies or bands for industrial, scientific and medical purposes are 915 M H z tor
0.915 G H z) and 2.45 G H z 1,2. A ll microwave ovens for home cooking operate at 2.45
G H z. Other frequencies, such as 20.2-21.2 G H z. 28 G H z. 60 G H z and 140 G H z. have
also been allocated.
Frequencies not allocated for industrial or consumer use are
employed in communications for radio and cellular telephones among other uses.
A
more detailed listing o f the allowed frequencies for worldwide industrial, scientific and
medical use is given in Table 2.1'.
Microwave technology can be dated to World W ar II2'4. when radar was first
developed for military purposes. After the war, several researchers such as G .B .B .M .
Sutherland explored the use o f microwaves to dry large ceramic whitewares. However,
it was not until the late 1960’s, w ith the filing o f the Levinson patent on firing ceramics,
that interest in microwave processing began to grow2. Table 2.2 shows the evolution o f
microwave technology2 over the years.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X-Rays
... rfttfn n d (
- Mcroweves
\*— L a w Radiation
io^'iT#
■
n*
iP
■
3 . 10'J
J ilO '«
i7»-----Z* — IP --- ^3— jp — Io^l-----,
Wavs Length (matin)
■
3*10*
‘
»
m p*
3k10*
Frsquancy (MHz)
■—
3B(10>
#WV\#
Molecular
vibrations
inner-shall
•lectrons
Outar-ahall
(valence)
electrons
Molecular
rotations
Figure 2.1. The Electromagnetic Spectrum and Response Mechanisms o f Molecules
(reference3)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Frequency
MHz
433-92
A rea perm uted
Frequency
tolerance
±
0-2 V*
Austria. Netherlands, Portugal,
West Germany, Yugoslavia,
Switzerland.
896
10 M H z
Great Britain.
915
13 M H z
North and South America.
2375
50 M H z
Albania, Bulgaria. Hungary.
2450
50 M H z
Worldwide (except 2375 M H z
regions).
3390
0-6 %
5300
75 M H z
6780
24150
40630
Netherlands.
Worldwide.
Netherlands.
0-6
125 M H z
Worldwide.
Great Britain.
Source: Metaxas and Meredith. 1983.
Table 2.1 The microwave frequencies allowed for industrial, scientific and medical use
(reference ')
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
9
rna'aOieiecinc properties ot ceramcsl, i
1960*1 •
Patent on flnng ceramics (196S1I3!
Formation ol tWPt (196fi)t<h*
M/w plasma sintering (1968)IS|
12ZQSApplications: drying, dewa*mg casting molds1*71
Melt mica, pyre*, porcelain (1970)1*1
Fire 300 us (13S kg) AtjOj castaOtes( 197S)W
neat A ljO j> t700*C (1976)1 "’t
Metnod to sinter relractones (1979)*"'
Melt U 02M
'IMPI • International Miuowave Power Inaiiute
1980*1.
Applications(pilot scale)
• Sipcastingi'1’1*!
• FiresparKplugftousingsi"'
• Clinkeringcementt'*!
Newareas (RS0>
. Giassi’7-"!
• Oxides, nonoxides, composites!"'
• Joiningandsealng!70*7')
MRS Microwave ProcessingSymposium
(1908)1“ *’’1
1990*1 •
Microwave ProcessingSymposia
(1990IW,1991IUI.1992IW)
Rapidproiileraiion(R40)i“ **'
• Newprocessapplications
• Newmatenals. microstiuctures
• Improvedheatingcontrol
• Modelnqandmeasurements
Table 2.2 The evolution ot microwave technology
(reference :)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
The extraordinarily rapid heating and high temperatures obtainable were first
reported by Haas et al.2 in urania gels in 1979. The first report on the same effects in
“white" ceramics (especially alumina and silica) was presented by Roy, Komameni and
Yang5 in 1985. What has been described in this section is the beginning o f microwave
processing.
More recent advances w ill be discussed in subsequent sections as they
pertain to this work.
2.2 Microwave-Material Interactions Theory
When electromagnetic energy is applied to a molecule, several response reactions
are possible as shown in Figure 2 .13. X-rays and ultraviolet rays (high frequency-short
wavelength radiation) cause electronic excitations in materials that are absorption
mechanisms for that applied energy. In the infrared region, molecular bond vibrations
are excited. The response mechanism to the applied field is the stretching, wagging and
bending o f molecular bonds.
In the microwave region, molecular rotations and
oscillations occur in the dipoles. Additionally, ions and electrons within the system can
migrate as a response to the applied microwave energy. These interactions give rise to
the two predominant loss (absorption) mechanisms in the microwave region: dielectric
loss and conductive loss.
One o f the mechanisms by which a material absorbs microwave energy is through
its dielectric properties6'7,8'9,10,11 or rotation o f dipoles. A material w ill couple with or
absorb microwave energy when it becomes “dielectrically lossy".
The degree o f
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
II
absorption is related to the material's complex permittivity ( e * ) \ Complex permittivity
is given by equation 2.1,
e*=E/-je"= E0(s;r-je;'etr)
(2.1)
where: ( eO is the real portion o f the dielectric constant
(e") is the imaginary portion o f the dielectric constant.
( j ) = H ) l/2.
(e>) is the relative dielectric constant,
(Go) is the permittivity o f free space
( e 'cit) is the effective relative loss factor.
As an electromagnetic wave o f proper frequency ( if the frequency is too high or too
low. dielectric measurable absorption w ill not occur12) propagates into a dielectric
material, currents are built up inside the dielectric material that attempt to match the
frequency and oscillation o f the incident wave as shown in Figure 2.2b . If the system is
a perfect dielectric (i.e.. no loss mechanisms) a charging current results that has a
maximum value when the incident wave is at a minimum.
charge storage.
This results in complete
A lossy material, which approximates real situations, w ill dissipate
some o f the energy due to resistance o f motion through friction, which gives rise to the
heating. This loss mechanism gives rise to a charging current that is somewhat out o f
phase with the initial wave. The tangent o f the angle between the incident wave and the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
< < < < < < < <
I V
/
H»
|
\w
^
\ .
O'
0-
^
^
/
/
i**
\
L
. K.
O ——
"V
0
/
V
b.
Figure 2.2. Response Mechanism o f Water to a Microwave Field when (ai the Field is
Applied and (b) the Field is Relaxed
(reference ;)
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
13
charging wave is an accurate measure o f the dielectric loss and is called tan 8, or the
dielectric loss factor2'613.
Dielectric loss factor (tan 8). which is directly related to
(e"C(t)6. can be described as the ratio o f the dielectric relative loss factor (c'dr) to the real
portion o f the dielectric constant (e~) as shown in equation 2.2.
tan 6=(e',cir) /e
(2.2)
The dielectric loss tangent value increases with temperature because the dipoles are
allowed freer range o f motion due to the thermal motion within the material.
heating
runaway
The
then increases exponentially as tan 8 increases; this is called ‘‘thermal
nJ.M.IS
Ionic and electronic motions are other methods o f microwave absorption. Ions and
electrons can also interact with the applied electric field and migrate.
This migration
causes a flow o f current that results in Joule heating ( I: R)31' 16,1 due to the resistance to
that flow. This situation occurs mostly at low microwave (Radio) frequencies1" and can
be described as a component o f the less tangent as shown in equation 2.3. where a is the
conductivity and f is the frequency.
tan 8= 0 /2 7 ^ 0 ^
(2.3)
Hence, the loss tangent can be used to describe both mechanisms o f loss.
Thus, the
power absorbed by the material, which provides the basis for the heating6, can be
represented by equation 2.4611. where P is power absorbed and E is the value o f the
internal electric field.
P=CT|E|:=2jtfE0E'r tanS|E|:
(2.4)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
Dielectric loss and ion migration are important and primary factors for microwave
absorption and the consequent heating o f dielectric materials.
What has been described above is the absorption o f microwave energy by dielectric
or insulating materials. Microwaves are also absorbed by metals3'6'11'1415. Conductors
do reflect most o f the microwave energy back toward the source o f the radiation: this
can be seen by Figure 2.36'16'17. However, some o f the radiation is absorbed up to the
"skin depth” . The skin depth (d) is the distance an electromagnetic wave may travel in a
metal or other conducting material before the field strength reaches 1/e or 0.368 o f its
value at the point o f interaction (which is at the surface o f the metal). Skin depth varies
with the square root o f the frequency as shown by equation 2.5. where p is the magnetic
permeability and f is the frequency.
d =I/(p far)1‘
(2.5)
Skin depth w ill also vary with the electrical conductivity (a ).
conductivity results in smaller skin depths as shown by equation 2.5.
Increasing
Hence perfect
conductors reflect microwave energy. The mechanism by which the metal is heated is
by currents induced by the electromagnetic field penetration.
These currents, called
"eddy currents”, build up a friction within the metal which causes joule heating.
Faraday's inductance law is a likely explanation for this effect: as the magnetic field o f
the electromagnetic wave propagates into a material, an electric current is induced that
acts perpendicularly to the applied magnetic field. This is also explained by M axw ell's
first field equation, which is a revision o f Faraday's law.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
M aterial type
Penetration
TRANSPARENT
(Low loss
Insulator)
Total
OPAQUE
(Conductor)
None
(Reflected)
ABSORBER
(Lossy Insulator)
Partial
to Total
ABSORBER
(Mixed)
(a) Matrix = low loss Insulator
(b) Fiber/particles/additives =
(absorbing materials)
Partial
to Total
A/VWWV
f \ / \ / \ A /V V v * -
Figure 2.3. Interaction o f Microwaves with Materials
(referenceJ)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
2.3 Microwave Processing
2.3.1 Review o f microwave applicators
Currently, there are several types o f applicators that are used in microwave
research and applications.
The most common applicators are the “single mode" and
“multimode” microwave units.
A “mode” can be described as the way an electrical
field is set up within the microwave cavity18.
Single mode microwave units utilize transverse electric (TEm,,. TEou. and TE m )
and transverse magnetic (TMoio. TMo:o and T M u n) modes9 1,119 :o. Single mode units
function by initiating a microwave signal through an aperture in a metallic tube. This
signal is reflected by a metallic structure that serves as a short circuit plunger located
opposite the aperture (Figure 2.4). The initial and the reflected waves form a standing
wave pattern that has a minimal electrical field value at the aperture and at the plunger.
The position o f the maximum value point o f the field can be fixed by the positioning of
the plunger.
As shown by Figure 2.4.
the sample is placed at the point where the
electrical field strength is at a maximum. In this way the coupling ability o f the sample
is enhanced.
The most popular o f the TE mode units are the TE|i)n and the TEou modes1*’.
Figure 2.4 is a TEion mode unit, where “n" is the number o f maxima within the cavity18.
The TEion rectangular cavity allows for a uniform electric field parallel to the axial
direction o f the sample. Figure 2.5 represents a TEou cylindrical applicator which has
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To rmcrowov*
M u re * via
a cvcu ta lo f
Ni n i c o n t ix t
vlMf
Cue ml pluoa*f
Coupling ins
S fC tan A* A
Covdy dulonct
Figure 2.4. Single Mode T E ,0n Microwave Applicator
(reference ” )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SUOINC SHORT
COAXIAL
INPUT PORT
MATERIAL
ADJUSTABLE
COUPLING
PROUE
Figure 2.5. Singie Mode TE,)I; Microwave Applicator
(reference
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
a special electric field pattern exhibiting a maximum field strength at the surface o f the
sample and a minimum at the sample center.
The TMoio unit, which is shown in Figure 2.6. is an applicator for processing thin
cylindrical samples.
The coupling in this unit is through the magnetic field in the
waveguide. In this type o f unit the aperture is fixed so that the cavity can only process a
certain type o f material10. Any changes in the material to be processed would require
changes in the cavity.
For this reason the T E mode units are convenient and more
utilized. The higher T M modes. T M 020 and T M n „ modes, allow for larger cavity sizes
and thus larger samples can be processed.
Single mode units are useful in processing materials that have small dimensions,
such as fibers, or materials that have low dielectric loss. Since the single mode cavities
are designed in such a way that a standing wave pattern results, low loss materials and
small materials can be placed in an area o f highest field strength to achieve
enhancement in their absorption.
M ultim ode cavities are the most popular microwave units.
They are most
commonly used in the homes o f consumers as ovens for cooking purposes. They are
also extensively employed for research and industrial processing tools.
Figure 2.7
shows the basic principle behind the multimode unit. The multimode unit, as the single
mode units, makes use o f a standing wave pattern caused by the in-phase oscillation o f
the incident wave (from the microwave source) and the reflected wave (from the
metallic enclosure comprising the cavity). However, at certain frequency ranges10 this
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
:o
Magnate field l«es
To microwave source
vo a circulator
r~
'
i
i
Coupling
_ !f 'V
Ceramic
material
fM0B C a vity
Figure 2.6. Single Mode TM.,|0 Microwave Applicator
(reference lJ)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cool'ng
A ir
-outcut
Co»>ly
S liiff'
M o g n *lio n
Input *101
W n f M esh lo
B » fv fn l m .c ro w o v r
ifo^ogf
/
H T T r o n tlo r m ff
Ooor la tc n *ioi a n d
v j l f l y lock m »chon«»m »
Figure 2.7. Multimode Microwave Applicator
(reference3)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
unit can support several modes, hence the name “multimode”. When the unit has no
load, or is empty, all excited modes show sharp resonant responses, shown as f t and f;
in Figure 2.8. When there is a coupling material or a dielectric load that partially fills
the cavity, the resonant responses o f the modes w ill follow f | and f \
The modes
broaden and shift frequencies, providing continuous coupling due to the overlap o f the
fields.
In theory, in a multimode cavity larger samples can be processed without
developing temperature gradients and non-uniform heating.
2.3.2 Review o f microwave sintering and processing
Most o f the work carried out in microwave processing o f ceramic materials in the
last decade involves sintering21 " • 23'24~5"G*7‘28 29 3031. Several authors have noted that
many white ceramics (i.e., metal oxides that do not possess ions with a magnetic
moment) can absorb microwave energy at elevated temperatures (around 800°C).
One
o f the major advantages o f microwave processing over conventional methods is the
greater sinter densities that can be obtained at shorter process times and lower
temperatures.
Several workers have shown21.22.23.24 ^
mjcrowave processing gives better
sintering densities than conventional methods at lower temperatures and in less time. V.
Fang et al.2' 28 reported that in the processing o f hydroxyapatite (H A p ). an important
biomaterial found in the bones and teeth o f living organisms, for all sintering techniques
used, microwave processed samples showed higher sinter densities in less time than
conventional methods. These findings are shown in Figure 2.9.
Microwave sintering
provided superior results even without the aid o f hot pressing or other time-consuming
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
c
D
> s
U
a
u
a
u
u
■*U
4)
m
fi
f
Frequency
Figure 2.8. Mode Set-up in a Multimode Applicator
(reference l0)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
M
HQO
u
IQOO
£
800
ICO
90
3
u
?
u
600
a
?
400
c
80
70
200
60
0
4
2
Time, nin
6
50
'Jrssn Qensity. X
(B)
(A)
Figure 2.9. Comparison o f Microwave and Conventional Sintering o f HAp
(A (-shows feature o f the thermal runaway in the microwave sintering; < I (-Thermal
runaway at 3 min. final temp=1200°C; (2 (-thermal runaway at 2.5 min. final
temp=1300°C
l B (-shows final densities with respect to green densities for HAp; (1 (-microwave to
1300°C for 10 min; (2)-microwaved to 1200°C for 10 min; (3 (-conventional to 1300°C for
2 hours; (4)-conventional to 1200JC for 2 hours
(reference :s)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
techniques.
When the green densities were higher this effect was even more
pronounced. This is attributed to intensive mass diffusion caused by uniform heating as
well as high heating rates27'28.
The trend o f higher densification at lower temperatures is also noted by M . Janney
and H. Kimrey for alumina.
Figure 2.10 shows data for alumina sintered in a
microwave field at various temperatures and times with the resulting densities.
In this
work alumina was sintered at 28 G H z as opposed to the standard 2.45 G H z. Alumina
couples at lower temperatures when higher frequencies are applied because its loss
tangent increases with frequency. Figure 2.11 shows a general relationship o f frequency
to loss tangent at room temperature.
Though Figure 2.11 shows the dielectric loss
profile o f water, the toss profile o f alumina peaks at higher frequencies and is a
minimum at 2.45 GHz.
The work by Janney and Kim rey29 supports the experimental results and data
presented by other workers. However, the precise mechanisms that lead to densification
enhancements in microwave processing as compared to conventional methods are the
subject o f much debate13. Additional work by Janney and Kimrey on the kinetics o f
microwave sintering reports that the activation energy for microwave sintering is 66%
less than by conventional sintering for alum ina".
The activation energy for alumina
sintering is 575 KJ/mol (conventionally sintered) and 160 KJ/mol (microwave sintered
at 28 G H z). Additionally, the microstructure is more uniform in the microwave case.
Figure 2.12 shows the Arrhenius plots for alumina sintered conventionally and at
28 G H z in a microwave applicator. This figure shows the log o f the sintering rate
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
Microwave
sO
0s
£
(0
c
©
O
Conventional
50 •u “
800
1000
1200
1400
Temperature, ’C
Figure 2.10. Microwave Sintering ot'Alum ina at 28 G H z at Various Temperatures
(reference ::)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30 MHz
300 MHz ,‘V > " "3 GHr ' r
fttqtnncv
..I—
-
.-
j . *
'
'*■
~
•*
*Sr -*
Figure 2.11. The Relationship between Frequency and Loss Tangent at Room
Temperature
(reference ;Q)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.0
*
8
0.5 :
0}
To
0.0 :
\
v
cr
OJ
c
'C
a)
J:
□
V,
cn
o Microwave
160 kJ/mol
□ Conventional
575 kJ/mol
O)
o
0.55
0.65
0.75
1000/T (1/K)
0.85
Figure 2 .12. Comparison o f Arrhenius Plots for Microwave and Conventional S interin
o f Alumina at 28 G H z
(reference ;:)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
plotted against 1/T (K ) for both methods. The slope o f the line is the activation energy
(E *).
Janney and Kim rey hypothesize that grain boundary diffusion is preferred over
surface diffusio n", due to the uniform heating that microwave processing affords.
Janney and Kimrey also attempted to diffuse trace amounts o f O 18 into single
crystal sapphire.
The Arrhenius plot is shown in Figure 2.13.
Janney and Kimrey
reported that the difFusivities were increased by 2 orders o f magnitude in a microwave
field in comparison to conventional techniques. The activation energy was reduced from
710 to 409 KJ/mol.
They attributed the diffusion enhancement to coupling o f defects
(vacancies) w ith the applied microwave field.
Katz et a l.13 investigated the diffusion o f chromia (C r:O j) into alumina at various
temperatures and conditions using microwave processing. Figure 2.14 is the Arrhenius
plot obtained for microwave processing compared to an Arrhenius plot made for
conventional attempts by Stubican and Oshenbach and by Oishi and Kingery. As can be
observed from the plot, there is no difference in the activation energies, only in the pre­
exponential factor.
The fact that both lines have the same slope indicates that the
activation energies are similar. The microwave data shows a factor o f 3 increase in the
intercept (pre-exponential factor) over the conventional case. This data consists o f only
three data points, hence finding a good curve fit would be difficult and the error
involved could be very large. Perhaps it might have been better to generate more data
points between the reported minimal and maximum temperatures to determine the plot.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
*18
«!
<N'
£
Q
9
-20
21
4.7
4.9
1
5.3
5.5
,7
(x 10"4)
i n ’ o /K )
Figure 2.13. Arrhenius Plot o f the Microwave-Driven Diffusion ot O ls into Single
Crystal Sapphire
(reference " )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Temperature
1750 1700
C
1600
-as
a
c
-26
/
-27
- Calculated from the data of
Stublcan and Osenbach, and
■ Oishl and Kingery
\
s
\
-20
0 .00040
------ 1------ 1------ 1------ 1------ 1----- 1------1----- 1
1
0 .0 0 0 5
0 .0 0 0 5 2
0 .00054
1 /T
0.00056
'-----r
0 .0 00 5 0
0.0 00 6
(K"1 )
Figure 2.14. Calculated and Measured Interdiffusivities for the Alumina-Chromia
System
(reference n)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3.3 Review o f microwave synthesis techniques
Unlike conventional sintering or synthesis, microwave synthesis depends largely
upon the ability o f one or more phases in the reactant mixture to couple with the applied
field. This may enhance reaction rates due to the fact that heating occurs throughout the
reactant mixture, rather than by heat conduction from the surface to the core as in
conventionally
processed material.
Hence, microwave heating
is a volumetric
phenomenon. Several compounds have been successfully synthesized in a microwave
f l e l d 3 :.3 3,34,35.36.37.J «,3 P .0 .4 M :.4 3 .4 4 ,5.46 .47.4«.49.S0.5..52.53
^
^
^
reactions32, polymer curing33, liquid state reactions and various organic reactions * have
been attempted along with solid state reactions. Though extensive amount o f work has
been reported in these areas, non-organic solid state reactions w ill be the focus of this
review.
M any types of solid state reactions have been attempted in a microwave field'11 and
many enhancements over conventional methods have been shown.
The reasons for
enhancements in reactions may lie in the increased mobility o f ions and rapid energy
transfer mechanisms from the coupling component to the bulkJ’Sj‘,'4°.
Boch et. al.4'’""0 reported making reaction-sintered mullite and aluminum titanate
from A hC b-SiO : and AUCh-TiO: materials respectively. Microwave processed mullite
gave 81% densification and 100% reaction after 30 minutes at 1580°C. Conventionally
processed material gave only 68% densification and 84% reaction product under the
same conditions.
In the case o f aluminum titanate. microwave processing showed
enhanced effects on sintered density and even more pronounced effect on reaction. At
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1250°C. which is near the lowest equilibrium temperature for the formation o f Ai:TiO<.
microwave processing showed reaction initiation.
Conventional processing shows no
reaction at 1250°C. A t 1290°C, microwave reaction showed 78% Al:TiCK formation,
whereas conventional methods did not produce any A b T iO f.
Lead barium titanate has also been successfully synthesized in a microwave field
by Cheng et al. from a solid state reaction using BaCCb. T iO ; and Pbj0 4 as precursors.
The synthesis was completed at 750°C in 10 minutes. Conventionally, the reaction is
completed at 850°C in 2 hours and usually is accompanied by PbO evolution51.
In the
microwave case, it was found that the volatilization rate o f PbO was much less than the
conventional case due to lowering o f reaction temperature and time. Additionally, it is
thought that i f the reaction rate is increased, this will result in the completion o f the
reaction before any Pb volatilization.
Zinc aluminate spinel (ZnA l:Q ») was microwave synthesized using ZnO and A L O .
by Ahmad and Clark43 and attempts were made to ascertain the reaction kinetics in the
microwave field and compare them with conventional processing. The results from this
study indicate that the microwave case showed enhancements in the reactivity and that
the kinetics path had changed.
Conventionally processed ZnALCb formation kinetics
was found to follow the Valensi-Carter model4’’ which is shown in equation 2.6. where
x is reaction extent. A is the change in volume as a ratio between reactants and product,
t is time and K.vc is the reaction rate constant.
K vc t= {A -[l + (A -1)x]“ 3- (A - 1) (1 - x r 3}/(A -1)=f(.x)
(2.6)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
34
The data from the microwave and conventional experiments are compared in
Figure 2.15.
Each condition (microwave or conventional), with respect to processing
parameters, was treated analogously. It was found that below 900°C , both microwave
and conventional reactions appear to be diffusion controlled, with ZnO diffusing into
alumina. This can be noted by the similarities o f the attempts to plot the Valensi-Carter
equation. The lines plotted at 900°C and 800°C both pass through the origin.
Above
900°C, neither microwave nor conventional techniques behaved according to the
Valensi-Carter model. This was due to the particle size ratio for the reactants being too
large.
This introduces geometric variables that the Valensi-Carter model does not
account for.
What can also be noted from the data is that the microwave case behaves very
differently from the conventional case, assuming the same problems with particle size
ratio exist with both cases.
The difference in reaction pathway is currently not well
understood. Enhanced diffusion is the current rationalization. However, the possibility
o f a different pathway or mechanism should not be discounted.
As in the case o f
sintering, there are several reaction mechanisms that are possible and multiple pathways
may be involved in one reaction process. Once a reaction pathway can be determined
for a conventional model, the same reactants and conditions should be used to
microwave process the materials.
Deviations from the parent kinetic model for the
reaction should be investigated by curve fitting the conflicting data to other kinetic
models.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
|4-|i*U-i)«l
-(*- *1<1-*)
!/(»-■>
35
0.4
0.4
O.J
O.J
0.2
0.2
0.1
0.1
0.0
0.0
A 1000 C
0
15
JO
45
60
0
75
REACTION TIME IN MINUTES
15
JO
45
60
REACTION TIME IN MINUTES
Figure 2 .15. Comparison ot'the Curve fining o f the Alumina-Zinc Oxide Reaction to the
Valensi-Carter Model in (a) Microwave and (b) Conventional Applicators
(reference >:)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
2.3.4 Review o f novel microwave applications and patents
In addition to sintering and synthesis, microwave radiation has found uses in the
areas o f liquid state processing, low temperature processing, polymer processing,
joining and organic material bumout. Some o f the novel uses o f microwave technology
include microwave dilatometry. hybrid heating technology, and microwave pressure
sintering.
In liquid state processing as well as in low temperature applications, the removal o f
water is primarily involved. The US Bureau o f Mines developed a technique for wet
chemical analysis by using microwave energy to heat materials to be analyzed in a
sealed vessel.
In this method, the materials (which may be minerals, ceramics, alloys,
etc.) are subjected to treatment with acids prior to heating'’.
The advantage o f
microwave application here is the allowance for an increase in temperature and pressure
that decrease digestion times from hours to a few minutes. In this work a microwavetransparent Teflon vessel is used.
The temperatures and pressures achieved in this
particular application were 180°C and 8.2 M Pa respectively.
Microwave radiation has also been applied to dry slip-cast materials2'6.
As the
particles are allowed to settle in a mold, the liquid layer can be driven o ff in a
microwave furnace. A fter the supemate is removed, the green body itself can be dried
and then fully fired. Due to the use o f microwave technology in this technique, the time
required for the entire process (from slip casting to complete firing) can be greatly
reduced, and efficiency and productivity are increased.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Further, microwave joining has become an energy-efficient method o f combining
two ceramic bodies at their interface6'54. In the joining o f alumina plates, a glass sealing
powder was used at the surface o f the plates. The powders were made o f a microwaveabsorbent material55.
This process allowed for the rapid heating o f the interface
material and thus a faster reaction during joining.
Microwave heating has also been applied to the burnout o f organic binder material
and polymer processing. In ceramic processing techniques, such as the consolidation o f
powders into a dense sintered body, binder materials are used to aid in obtaining various
shapes for ceramic green bodies. Microwave furnaces have been successfully used in
the removal o f these organic materials or binders6. In this process either the microwave
energy couples w ith the organic binder itself to give bumout or. if the ceramic material
is more lossy than the organic, it couples with the ceramic phase which heats the organic
and causes bumout.
Novel uses o f microwave energy involve dilatometry56 and the combination
microwave/conventional
hybrid
furnace.
Dilatometrv
basically
involves
measurement o f material shrinkage or expansion as a function o f temperature.
the
The
information obtained is used in determining the maximum sintering density o f a
material and the sintering rate.
Most dilatometers use conventional heating methods
and are the standard tools for shrinkage or expansion determination.
dilatometers are normally
incompatible
Conventional
for microwave application due to their
constituents'6. The microwave-compatible device involves the use o f a microw ave
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
applicator with a provision for sample placement. The sample inside the applicator is
connected to the dilatometer located (Figure 2.16) outside the microwave applicator.
The sample is connected to the dilatometer via an alumina pushrod (alumina is
relatively transparent to microwave energy at low temperatures due to its low dielectric
loss).
The rod is allowed to enter the microwave cavity through a small aperture,
approximately
1/4 o f the microwave wavelength, that is designed to allow
no
microwave leakage outside the cavity.
The sample temperature is recorded by an optical pyrometer stationed outside the
furnace in a small port adjacent to the sample on the outer wall o f the cavity.
As the
sample is heated and begins to sinter, the push rod (which is in intimate contact with the
sample) follows the shrinkage which is recorded by the dilatometer.
The
combination
microwave/conventional
developments in microwave technology.
furnace5,5*
is
one
of
the
latest
It combines the properties o f both heating
methods to give w’hat is known as “microwave hybrid heating” (M H H )54.
This is
accomplished in a microwave-only furnace by use o f a material with a high dielectric
loss at relatively low temperatures (secondary coupler) which surrounds the sample
(which has little or no loss at low temperatures) to be sintered or processed.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
Figure 2.16. The M icrowave Dilatometer
Designations: (20)-microwave furnace. (30)-dilatometer. (40)-optical pyrometer. (
sample under test. (60)-measurement area, (70)-alum ina pushrod. (80)-aperature. (90)linear variable differential transducer. (lOO)-contact surface
(reference 56)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
40
The secondary coupler absorbs energy first and then radiates that energy in the form
o f heat to the sample.
This continues until the sample’s temperature approaches the
point where the sample begins to absorb microwave energy.
W ith proper choice o f
materials, at about that point the secondary coupler ceases to absorb and the sample is
the source o f heat generation.
The combination furnace (shown in Figure 2.17) effectively accomplishes M IIH .
without the need o f a secondary coupler. It provides a conventional furnace which has
heating elements that are located on either side o f the processing cavity in a secondary
chamber. The process cavity, which is contained within the secondary chamber, has a
wall which allows very little microwave leakage from the system. This wall, as is most
o f the system, is made ofaustenitic stainless steel because o f its conductivity (thus high
reflectance o f electromagnetic radiation), good welding characteristics and oxidation
resistance. The walls o f the process chamber are perforated (4m m holes) on both sides
to allow entry o f conventional heat from the elements in the secondary chamber. The
secondary chamber can maintain a temperature o f 1200°C.
The rear o f the process
chamber contains a port which is the end o f the wave guide for the microwave
applicator. This setup allows for conventional preheating, followed by the microwave
processing o f materials, thereby eliminating the need for a secondary' coupler and "hot
spots". Hot spots can develop in the secondary coupler and allow the secondary coupler
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
41
to continue coupling even after the sample becomes lossy.
This happens because the
hot spot is the most lossy part o f the system, due to the fact that its temperature may be
several hundred degrees higher than any other part o f the system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
■■26
/•
Figure 2.17. The Combination Microwave-Conventional Furnace
Designations: (lQ )-kiln . (20)-rear o f chamber, (22)-perforated walls, (24)-secondary
chamber, (26)-heating elements (conventional), (28)-microvvave port, (30-31 )-microwave
waveguide section, (34-36)-insulationing and outer housing
(reference 57)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
Chapter 3
E X P E R IM E N T A L
3.1 M icrowave equipment
The microwave applicators or furnaces are the only equipment described in detail
in this chapter. The conventional units and equipment are described in less detail in the
subsequent chapters.
3.1.1 Panasonic 900 W multimode microwave furnace
One o f the furnaces used in this work was a multimode 900 watt (maximum output
power) unit operating at 2.45 G H z. The microwave cavity was approximately a cubic
foot in volume (14” x 14.75" x 8.25").
A small hole was drilled at the bottom o f the
furnace for the insertion o f a thermocouple.
3.1.1.2 Temperature measurement
Type-S (Pt.10% Rh) thermocouples were used for temperature measurements. The
thermocouples were made by arc*welding the ends o f
rhodium wire.
platinum and
platinum-10%
The non-welded ends were then guided through small openings o f an
alumina tube for thermal shielding.
3.1.2 F iv e-K W multimode unit
This unit was used exclusively for PZT synthesis because PbO vapor could be
vented easily due to the unit's cylindrical cavity. The top o f the cavity was vented into a
hood placed over the unit. The unit has a variable power capability which can adjust the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
44
microwave power from the water-cooled magnetron during any part o f an experiment.
The cavity surrounds a quartz tube which is the process area for the sample. The sample
is placed on alumina fiber and it is in contact with a thermocouple, which is introduced
from the bottom o f the cavity.
3.1.3 M M T Model 10-1300
The microwave furnace used for most o f the synthesis work in this study was a
M icrowave Materials Technologies (M M T ) model 10-1300 ( M M T Inc. Oak Ridge.
T N ).
This multimode unit. Figure 3.1. was capable o f supplying 1300 watts o f
maximum output power.
The cavity is made o f stainless steel.
An antenna, which
distributes the field, is located on the floor o f the cavity. Above the antenna, the unit is
equipped with a glass plate which is used for placing materials for processing. The unit
body also has three apertures that can be used for inserting a thermocouple. Two o f the
apertures are located on the back plate o f the cavity, and the third is located on the top
plate o f the cavity. These apertures can be equipped with swagelokrM-type fittings that
can be tightened firm ly around a thermocouple housing to ensure that the thermocouple
housing is grounded to prevent interactions with the applied electromagnetic field.
Additionally, there is an aperture that can be used for monitoring the temperature using
a pyrometer. This is located on the side wall opposite the magnetron assembly. Finally,
there are two additional apertures on the bottom o f the side wall opposite the magnetron
which can be used for the introduction o f a process gas into the chamber. This function
is in use when the atmosphere chamber is operating (described later).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
45
The model
10-1300 uses a Universal
Digital Controller model
UDC
3000
(Honeywell. Inc.). This is a programmable controller which controls temperature via a
thermocouple that is placed in contact with the material to be processed inside the
microwave unit. The controller reads the em f sensed by the thermocouple and rectifies
the power supplied to the magnetron to maintain a programmed heating rate or set point
temperature.
The model 10-1300 is also equipped with an atmosphere chamber (Figure 3.2).
which can be used to maintain a process gas atmosphere during experiments.
The
chamber has a Teflon gas inlet line and aluminum outlet line. The inlet line is connected
to a flow meter outside the chamber for gas regulation. The outlet line is connected to a
vacuum pump and an exhaust hood. The atmosphere chamber consists o f a quartz bell
jar that is mounted on a Teflon O-ring
The ja r is placed on an aluminum baseplate. A
second O-ring is placed over the jar such that it fits above the rim o f the jar.
An
aluminum ring is placed over the jar and over screw posts, where the aluminum ring is
secured by using screws.
The chamber unit is electrically grounded by fastening the
aluminum gas outlet line to the outlet aperture o f the cavity.
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
46
Figure 3.1. The Microwave Materials Technologies Model 10-1300 Microwave Unit
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
Quartz
Cover
Alumina
Cover
Sample
[Alumina
Therm ocouple
Insulation
[Alumina
Cylinder
Aluminum
Baseplate
Figure 3.2. Schematic o f the Atmosphere Chamber for the Model 10-1300 Unit
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
3 .1.3 .1 Microwave supporting materials
A ) Insulation
During each microwave experiment, insulation was used around the sample to
provide secondary coupling or thermal insulation for the sample.
schematic setup for microwave processing.
Figure 3.3 shows a
The sample is placed at the center o f the
cavity, and is surrounded by hollow alumina beads (Orpac. Inc.). which arc used as
thermal insulation. The thermocouple contacts the sample from the bottom in the case
when a desired atmosphere is used, and when the atmosphere chamber is not in use. the
thermocouple is introduced through the top aperture.
The thermocouple is normally
sheathed in a sintered alumina cover, so that no reaction between the sample and the
thermocouple may take place.
In this work silicon carbide (S iC ) rods were used as secondary couplers and were
usually installed around the sample 1/2 inch away from the sample surface.
The SiC
was also surrounded by hollow alumina beads. The outer board, which surrounds the
alumina beads, was made o f either alumina fiberboard. Z A L-45 (Zircar. Inc.) or alumina
cylinder, A L C (Zircar. Inc.). depending upon the particular experiment. This served as
additional thermal insulation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
Low Density
Insulation
Secondary Coupler or
Insulation
Sample Under Test
T a b le B a r rie r S y s te m
Alumina Flbertjoard
Quartz Rings
Figure 3.3. Typical Set-up for Processing in a Microwave Furnace
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
3.1.3.2 Temperature measurement
Thermocouples have long been used in microwave experiments as a method of
temperature determination.
They are effective as long as they are contained in
conductive sheaths that can be grounded to the cavity.
This is done to prevent
electromagnetic interaction with the metallic thermocouple bead.
I f the unsheathed
thermocouple bead is exposed to the field, erroneous electrical signals are sent to the
e m f detector,
resulting
in
incorrect and
erratic
temperature
readings.
I f the
thermocouple is sheathed but ungrounded, electric arcing w ill occu; between the
thermocouple bead and the metallic sheath resulting in superheating, and possibly
melting, o f the bead.
The thermocouples used in this study were manufactured by
Omega. Inc.. and were supplied by M M T .
Type-K. (chromel-alumel) thermocouples
were used for low temperature studies (<1100°C ) in air and nitrogen environments.
Tvpe-S ( P t-P t/10%Rh) thermocouples were used for the high temperature studies
(7 0 0 °C -16 00 °C )
in
air
and
inert
atmospheres.
Type-C
(Tungsten-Rhenium)
thermocouples were also used for high temperature studies (>1000°C )
in inert
atmospheres.
For
all
non-atmosphere
chamber
experiments.
M o S i:-sheathed
type-S
thermocouples were used.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
51
3.2 Chemical preparations
A ll information on the reactants and precursors used, and their proper chemical
stoichiometries, are provided in each chapter for the corresponding material syntheses.
3.3 M aterial characterization techniques
The following section deals with the methods o f materials characterization used
during this work. Standard characterization techniques w ill not be described in detail.
3.3.1 X R D (Powder and high-temperature)
The principal method for phase characterization used in this work was powder X ray diffraction (X R D ).
and
microwave
A ll powders were characterized before and after conventional
processing
to determine
product and reactant
amounts,
phase
composition, and reaction pathways and kinetics. Several samples were run to reduce
the possibility o f preferred orientation. X R D can determine concentration level o f less
than 1% depending on sample preparation and scan speed.
High temperature X R D (H T X R D ) was used to identity’ the phases forming during
heating in situ.
A platinum hot stage was used for all experiments.
Experiments
performed in air were limited to a m axim um temperature o f 1200°C. Those attempted in
inert atmospheres were limited to a maxim um temperature o f 1300°C and those in a
vacuum could be conducted up to 1500°C .
The temperature differences in different
atmospheres were due to the limitations o f the equipment.
Hot gases could cause the
camera lens to implode during the cooling process.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
X R D was employed in this work show qualitative comparisons in the samples
processed under different time and temperature conditions for a given material system.
Several X R D analyses were made using different portions o f a sample to determine
average concentrations and to minimize preferred orientation effects. Hence the relative
peak intensities reported in this work for a given system are the average o f the peak
areas for each analysis. The analyses for a given sample are not normalized nor directly
quantified with the other analyses made on different samples. The reported X R D data
on a given material system shows trends the disappearance and appearance o f observed
phases as process time and temperature increase.
3.3.2 Dielectric measurements
In order to understand better the real time behavior o f the microwave reactions in
this work, dielectric measurements were performed at microwave frequencies.
This
technique was used to determine the real ( £) and imaginary ( ? ) components o f the
dielectric constant o f reactions attempted, as well as the dielectric loss factor (tan oi.
This study was very important because currently there is no technique available to study
microwave synthetic reactions in real time.
D T A and other thermal methods are
typically used to study conventional reactions.
However, due to the nature of
electromagnetic processing, there is no way to use a reference sample (as used in D T A
and D S C ) to compare with the heating o f the actual sample. This is due to the fact that
once a sample couples with the applied field only that sample w ill efficiently absorb.
The second sample will be heated by thermal conduction from the first sample.
This
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
situation would obviously lead to uneven and uncontrollable heating o f both the
samples.
Figure 3.4 is a schematic o f the apparatus used to determine dielectric properties by
cavity perturbation technique. A conventional furnace is employed to heat the sample
and the sample holder.
After heating to a desired temperature the sample is rapidly
moved from the furnace and placed into a high electric field region o f a thick-walled
cavity.
The resonant field and loaded Q values are measured by a Hewlett-Packard
8753 network analyzer.
Afterwards the sample can be transferred to the furnace for
subsequent measurements or allowed to remain in the cavity and measurements could
be made during the cooling cycle. 3.3.3 T G -G C -IR measurements
Thermogravimctric (TG)-gas chromatography (GC)-infrared spectroscopy (IR )
measurements were made on selected samples that decompose during reaction and emit
gaseous products. The sample was heated on a microbalance, to record weight changes
during heating. The o ff gases were carried into a GC for separation into their individual
components, after which each component was analyzed by IR for its composition. 3.3.4
Electrical conductivity measurements
Electrical
conductivity measurements were made on the non-stoichiometric
powders synthesized in this study. Conductivity measurements were made by placing
each powder to be analyzed between two conductive plates according to A S T M Test
Method number D 257. The weight o f each sample was closely monitored such that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
OFHC
*r
IMPUl
OUIP'JT
L*J
C A B lIS
u
r T j
tp'.UMAr
INSULATION
.4. Schematic o f the Furnace and the Hewlett-Packard Network Analyzer used
for Obtaining Dielectric Measurements
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
55
compartment volume was known. This is a requirement for conductivity measurement
validity, since fluctuations in the volume can affect a material’s conductivity.
3.3.5 Microstructure examination
Microstructure examination was carried out by scanning electron microscopy
(S E M ). SE M was used to determine grain size and for qualitative elemental analysis at
the micron-size level (using the backscatter mode or ED A X).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
Chapter 4
M IC R O W A V E S Y N T H E S IS OF A L U M IN U M T IT A N A T E
4.1 Introduction
4.1.1 General
Alum inum titanate (A liTiO s). also called tieilite. is the only thermodynamically
stable binary oxide in the A liO j-T iO i system60. The phase diagram is shown in Figure
4.1. It exists in two allotropic forms61: the high temperature form. a -A L T iC k which is
stable from 1820 to 1860°C. and the low temperature form. P -A b T iC k
not stable at room temperature, regardless o f the quenching technique.
the pseudobrookite structure, which is shown in Figure 4.2.
a-A LTiC K is
P-Al;TiO< has
Several compounds are
known to have this structure which is named after the mineral, pseudobrookite
(FeiTiO?).
Pseudobrookites have the general formula o f A *J; T f 4X'*< or A *: T f 4; X ': 5.
Fe. Ga or A l can occupy the A * 3-site. M g, Fe or Co can occupy the A *: -site6~.
From the phase diagram, A liTiC b is seen to congruentlv melt at about 1860°C.
However, upon equilibrium cooling between 1300°C and 750°C. AUTiO? is found to
dissociate into its constituents, T iO ; and
A I2 O 3 63.
This aspect was first investigated
b y
Lang et al64, who thought it to be a dynamic equilibrium competition between formation
and decomposition reactions o f A L T iC k
Several researchers have now confirmed this
phenomenon65 66-67 and identified equilibrium temperature range o f 1200°C to 1380°C.
depending
on
the
purity and on
the
particle
size o f starting
powders
used.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
57
2015'
2000
/
I860*
/
1820'
Liquid
1900
Liquid
I860'
Liquid
300
1820*
TiO
A !,0
Liquid
1700
A l,0
20
40
60
80
TiO
Figure 4 . 1. Binary Phase Diagram o f the System A l:0 3- T i 0 ;
(reference,(J)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
i
A
0 * “ Ory<^n «n
o * - C jfa u
Figure 4.2. The Crystal Structure o f p -A K T i0 5
(reference61)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S')
4.1.2 Applications o f aluminum titanate
This compound has found use mainly as a low bulk thermal expansion material.
Pure A liTiO s has a highly anisotropic thermal expansion.
The thermal expansion
values reported show shrinkage along one axis and expansion along the other two axes (
a=-3.0 x lO -6 K '1, b = l 1.8 xKT 6 K '1. c = 2 1.8x1 O'6 K ' 1 . calculated at 1 0 2 0 °C f8. Hence
this results in a net low bulk thermal expansion, and microcracking in its ceramic bodies
when cooled from sintering temperatures.
As a ceramic body. A liT iO j has low thermal expansion properties because o f a
complex system o f internal stresses and microfractures present62.
These stresses are
developed upon cooling from high temperatures and are due to the anisotropic behavior
o f the individual grains that comprise the body. This leads to the microcracks that show
opening and closing hysteresis upon heating and cooling 62 60. This also contributes to
the material having apparent low thermal expansion.
Low thermal expansion is made
possible by the microcracks or voids allowing an area for expansion during heating, so
the overall volume remains relatively unchanged.
This characteristic, along with the
relatively high melting point, makes it attractive in applications such as exhaust liners
and molten-metal filter materials70. However, due to the existence o f the microcracking
phenomenon, aluminum titanate does not have very good mechanical properties '. The
reported strength values are typically between 5-50M Pa61'71. depending on purity and
particle size o f the starting powders, as well as the sintering conditions. However this
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission
60
system is an interesting example o f a commercial ceramic which could be synthesized
by a new synthesis processes.
4.1.3 Synthesis techniques
4 .1.3.1 Conventional synthesis o f aluminum titanate
Alum inum titanate (A I:TiO s) can be made from A l; 0 3 and T iO : bv the following
reaction72'73'74'75:
A 1:0 3 + T iO : = A l:T i 0 5
Figure 4.3 is a typical X R D pattern for P-A l:TiO <. A l:T i 0 5 has been synthesized by
several methods including solid state reaction73 74,75 and sol-gel processing 6 '7.
Each
technique has advantages and disadvantages for A l:TiO < synthesis.
Complete solid state synthesis using T iO : and A l:O j without dissociation occurs
conventionally at temperatures above 1300°C w ith various soaking time periods,
regardless o f particle size7j 74,75. Equimolar amounts o f T iO : and A I : 0 3 arc wet-milled
for 10 hours or more to ensure maximal mixing and particle contact4. The mixture is
then heated to reaction or sintering temperature by several techniques, depending upon
the desired use o f the resulting material.
A lum inum titanate has been synthesized from a sol-gel process by several
workers76,77. Woigner et al. used tetrabutylorthotitanate and aluminum sec-butoxide as
the sources o f T iO : and A l: 0 3 respectively. The reaction involved co-hydrolysis o f the
reactants to give a homogeneous gel and particle sizes in the 60-200 angstrom range.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
9068
20
90
Figure 4.3. X R D pattern o f p-AUTiO,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The reactants were dissolved in 2-butanol and refluxed at 100°C for I hour in nitrogen.
Hydrolysis was carried out at room temperature by either moist air or dilute acetic acid.
The solution was poured on a glass plate and fast dried. The resulting xerogel was an
amorphous powder which was used to fabricate ceramic bodies by hot-pressing routes.
The A liT iO ; synthesized from this route shows a reaction mechanism similar to
solid state synthesis.
A ^TiO s, upon heating, does not form until 1300°C and the
reaction was complete in 2 hours at 1350°C. The grain sizes reported were in the 0.51pm range and the strength and Young's modulus were reported at 160M Pa and
lOOGPa respectively. The reported thermal expansion was 80 x 1 0 '7 °C . Hence
smaller grain sizes lead to higher thermal expansion values'7.
4.1.4 M icrowave synthesis o f aluminum titanate
M icrowave reaction sintering o f Al^TiOs has been attempted recently by Piluso et
al.49 and Boch et al.50. They found that A liTiO ? could be microwave synthesized by a
solid state reaction o f A l;0 } and T iO : beginning at 1210°C. as compared to 1300°C
conventionally.
The microwave samples showed about 93% o f theoretical density- at
1290°C whereas conventional sintering gave only 89% at 1330°C.
Hence the authors
demonstrated that the temperatures necessary to achieve high densities could be reduced
by microwave processing. These findings were the stimulus for the present attempt to
microwave synthesize AhTiOs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6)
4.2 Experimental procedures
4.2.1 Alum ina-titania (anatase) system
A lum inum titanate was prepared in the following fashion. Equimolar mixtures o f
alumina (A I 2O 3) (lot A16SG, Alcoa Inc), with a 0.2(im ave. particle size, and anatasephase T iO i (Cerac, Inc), with a <5pm ave. particle size, were wet-milled in deionized
water for 16 hours using A I 2O 3 m illing media.
The container used for milling was a
Nalgene™ bottle with a capacity o f 1 liter. The charge roughly filled 85% o f the bottle.
The m illing speed was about 65 revolutions per minute (rpm).
The media were separated from the reactant mixture through a 40-mesh screen.
The mixture first was dried in a microwave oven, and then in a drying oven overnight at
120°C.
In order to increase particle-to-particle contact, the samples o f the reactant mixture
were pressed into pellets, uniaxially at pressures o f 9000-1OOOOpsi.
The pellets had
diameters o f 2.873cm and heights o f about 0.76cm for all microwave and conventional
experiments.
4.2.2 High temperature X R D analysis o f the anatase-alumina reaction
Samples o f the A l 203 -T i 0 2 mixture were analyzed by high-temperature X R D to
identify phase transitions and reaction temperatures.
The samples were placed on a
platinum stage at room temperature, and a starting X R D pattern was obtained. Several
experiments were attempted to determine approximately where phase transitions and
reactions occurred.
Once temperature ranges were identified for a particular thermal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
event, the subsequent experiments were carried out at slow heating rates such that the
temperatures for transitions and reactions could be accurately determined.
4.2.3 Conversion o f anatase to defective rutile
Anatase-phase TiO : (Cerac, Inc) was placed in an alumina crucible and fired to
1100°C under a forming gas (5% hydrogen-95% argon, Linde Inc.) atmosphere for 3
hours to give oxygen-defective or Magneli phase rutile78'79. A fte r firing, the product
phase was black colored, and rutile was usually the only phase identified by (X R D ).
Whether the black phase has oxygen vacancies, as is usually assumed with no
supporting data, or Ti interstitials, as concluded by Porter80, makes no difference for
this study. The defective rutile phase is reasonably electrically conductive.
4.2.4 Alumina-defect titania (rutile) system
The defect-TiO: powder was dry-milled with alumina for 16 hours using alumina
milling media. Dry milling was used to prevent possible oxidation o f defect rutile in the
presence o f water.
The mixture was milled in a 1-liter Nalgene |TM1 bottle.
Pellets,
having the same dimension described in section 4.2.1, were uniaxially pressed from the
resulting powder mixture at pressures in the range of 9000-1 OOOOpsi with a Carver
Laboratory press.
4.2.5 Conventional synthesis o f aluminum titanate
Each pellet was heated on a bed o f alumina powder in an alumina crucible. One set
o f samples was fired in air in a Rapid Temperature furnace (C M Furnaces. Inc.) with
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
M o S ii heating elements. Another set o f samples was fired in nitrogen in a tube furnace
with M oS i; heating elements or an Astro Inc. furnace with graphite heating elements.
Samples were fired isothermally at 1250°C, 1300°C, 1350°C» I400°C and 1500°C
for
15, 30, 45, 60 minutes, respectively.
Four samples were heated at each
temperature/time period. The heating rate was 20°C/m in for all samples.
was allowed to cool as the furnace cooled.
Each pellet
After each experiment the samples were
characterized by X R D for phase composition.
4.2.6 M icrowave synthesis o f A hTiOs in air and nitrogen
Each pellet was placed in alumina housing surrounded by A I 1O 3 beads inside the
quartz atmosphere chamber o f the M M T 10-1300 microwave unit described in detail in
Chapter 3. A ir atmosphere or stoichiometric experiments required the use o f 3 SiC rods
as secondary couplers. Before the start o f each experiment the atmosphere chamber was
evacuated and backfilled with the desired atmosphere 3 times to ensure proper
atmosphere control.
For stoichiometric experiments, 80% o f the total power was applied to initiate each
experiment due to the inability o f T iO : and AI1O 3 to couple at low temperatures. The
heating rate was programmed automatically at 20°C/m in starting at 50°C below each set
point temperature to minimize the possibility o f overshooting the set point temperature.
The same soak times were used for the microwave and conventional experiments.
However, the soak temperatures used in these experiments were 950°C ,
1050°C.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
1100°C, 1 150°C, 1200°C. 1250°C and 1300°C.
Each sample was allowed to cool
without quenching and characterized by X R D for phase composition.
4.2.7 M icrowave synthesis o f aluminum titanate from reduced rutile
Each pellet was placed in alumina housing surrounded by A I 2O 3 beads inside the
quartz atmosphere chamber o f the M M T 10-1300 microwave unit described in detail in
Chapter 3.
N o secondary coupler was used for the defect-TiO; experiments.
The
vacuum and backfilling procedure used were the same as described in section 4.2.6.
Reduced rutile experiments required only 60% o f the total power, since coupling o f
the material was very effective. In these experiments, achievement and reliable control
o f a set point temperature was not trivial because o f the thermal runaway conditions.
The power output had to be manually adjusted for most experiments to achieve a
desired temperature.
The
maximum
soak time used in these experiments
temperatures 600°C , 700°C, 800°C and 900°C .
was
15 minutes, at
The samples were allowed to cool
without quenching and characterized by X R D for phase composition.
4.3 Results and Discussion
4.3.1 High temperature XR D analysis o f the anatase and alumina reaction
High temperature X R D (H T X R D ) information in Figure 4.4 shows that in a
conventional furnace aluminum titanate is not synthesized from anatase. but only after
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
AjJ-LkJ.
Figure 4.4. X R D Patterns for A1-0-. and TiO- heated at various temperatures
Designations: A l-T iO ;-0 ; A l-O ,-
9
: TiO - (anatase.)-,*.: TiO - (ru tile)- | :Pt stage- Q
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
the anatase had converted to the rutile phase.
Anatase begins to convert to the rutile
phase between 9 0 0 -l0 0 0°C in air. and near 800°C in inert atmospheres or in a vacuum.
A n A lfTiO s phase was not observed even at 1200°C when heated in air for 4 hours
(the instrument could not be operated above 1200°C in air). In an inert atmosphere or in
a vacuum, the reaction began between 1300-1400°C and was not completed even by
1500°C after 1 hour, which is the maximum working temperature achievable by the
H T X R D instrument.
4.3.2 Conventional and microwave A hTiO s synthesis in air
Representative powder X R D patterns for AIjTiO s, T iO : and A fC ^ from this work
are given in Figure 4.4 with the corresponding peak identifications.
Due to the very
large volume o f X R D patterns obtained in this research project, most data w ill be
presented in tabular form using peak intensities to display trends in the various reaction
pathways.
Hence, Tables 4.1 and 4.2 summarize the X R D findings o f experiments to
synthesize AhTiOs from ct-alumina and anatase in air by microwave and conventional
methods, respectively.
Both tables list the relative peak intensities for the strongest
representative peaks o f the reactant and product phases.
Table 4.1 shows that A liT iO s begins to nucleate at 1150°C using microwave
heating. A fter 45 minutes at 1300°C the microwave reaction is complete. Synthesis by
conventional methods does not show any product formation until 1300°C.
findings are consistent with the work o f Ph. Boch et al.50 cited earlier.
aluminum
titanate
nucleated
near
1210°C
in
microwave
and
These
In that work
near
1300°C
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
69
conventionally.
In the microwave synthesis study presented in this thesis, the reaction
temperature was lowered by another 50°C.
Table 4.1. X R D Analysis o f the Microwave Synthesis o f AhTiO? in A ir
Conditions
________Phases Identified by X R D and Relative Intensities
Temp/Soak
01-A I 2O 3
T i 0 2 (anatase)
T iO : (rutile)
1050°C/30 min
71
—•
100
—
1 l5 0 °C /3 0 min
42
-------
100
( 1)
1250°C/30 min
26
---
100
38
1300°C/15 min
19
—
96
100
1300°C/30 min
4
---
36
100
1300°C/45 min
—
---
A l 2T i 0 5
100
Table 4.2. X R D Analysis o f the Conventional Synthesis o f Al^TiOs in Air
Conditions
Phases Identified bv X R D and Relative Intensities
Temp/Soak
01-A I 2O 3
T i 0 2 (anatase)
T iO : (rutile)
Al:TiO<
1300°C/30 min
31
—
100
—
1300°C/45 min
32
—
100
6
1400°C/45 min
38
----
86
100
----
7
100
1500°C/45 min
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
4.3.3
Conventional and microwave A^TiOs synthesis in nitrogen
Tables 4.3 and 4.4 show the X R D results o f A^TiOs synthesis from alumina and
anatase in nitrogen by microwave (Table 4.3) and conventional (Table 4.4) methods.
The relative peak intensities indicate composition at each temperature and time interval.
Table 4.3 X R D Analysis o f the Microwave Synthesis o f A^TiOs in Nitrogen
Conditions
Phases Identified by X R D and Relative Intensities
Temp/Soak
CZ-AI2O 3
T iO ; (anatase)
T iO : (rutile)
A l: T i 0 5
950°C /15 min
52
7
100
—
9 50°C /30 min
46
5
100
1050°C/15 min
55
---
100
1
1050°C/30 min
49
--
100
j
1100°C/15 min
57
--
100
23
1100°C/30 min
37
---
100
34
Table 4.4 X R D Analysis o f the Conventional Synthesis o f A hTiOs in Nitrogen
Conditions
Phases dentified by X R D and Relative Intensities
Temp/Soak
0 A I 2O 3
T iO : (anatase)
T iO : (rutile)
A l 2T i 0 5
1150°C /30 m in
47
3
98
---
1250°C/30 min
41
---
100
---
1300°C/30 min
35
---
100
( 1)
1400°C/30 m in
18
---
93
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
It is clear from the X R D data that in an atmosphere that promotes oxygen defects
and reduction o f titania, A l2T i 0 5 can be synthesized at substantially lower temperatures
in a microwave field. In comparing the data from Tables 4.1 and 4.3 it can be observed
that the nucleation o f A l2T i 0 5 is lowered by at least I0 0°C by using reduced T iO ; .
However, a comparison o f Tables 4.2 and 4.4 shows that in conventional synthesis the
use o f reduced rutile does not affect the onset o f reaction.
This can further be corroborated by the dielectric data*1'82 acquired at microwave
frequencies in both air and nitrogen atmospheres using the same starting reactants, i.e.
alumina and anatase. The dielectric measurements quantify the "real part” (&) and the
"imaginary or loss part” (&') o f the dielectric constant at 2460 M H z as temperature
ttI jp
increases * .
From the data shown in Figure 4.5, the reactants heated in air up to 1400°C show
an increase in & and &/ starting at about 250°C. This rise peaks at 600°C and is possibly
due to water evolving from the test sample. The increase in £• and &•' starting at 800°C is
consistent with the phase change o f anatase converting to rutile. confirmed by H T X R D
(Figure 4.4).
The sharp rise in & and &/ starting at about 1160°C is likely due to the
nucleation o f A l2T i 0 5; this is confirmed by X R D results (Table 4.1).
Hence, in this
case, dielectric measurements can be used as an in situ technique to follow the reaction
in the microwave region.
Figure 4.6 illustrates the dielectric data (e'and e ') for the alumina-anatase reaction
in a nitrogen atmosphere. The data can be interpreted analogously to Figure 4.5. with
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
respect to phase changes; however, what is important to note is that the onset o f the
A I:T i 0 5 formation is at about 1050°C. This information is consistent with X R D results
obtained (Table 4.3).
The dielectric measurements demonstrate clearly that defect generation caused by
nitrogen processing lowered the onset o f reaction by about 110°C as opposed to air
processing.
Additionally, the increase in &/ starting at 250°C is greater in magnitude
than in the air measurement case.
A value o f about 1.2 is observed for the nitrogen
case and about 0.3 is observed for the air case.
In the air atmosphere case, this peak
decreased by 800°C; after that the peak for the phase transition o f anatase to rutile
begins. However, in the nitrogen atmosphere case, the 250°C loss-peak decreased to a
m inim um by 1000°C and probably includes the phase transition.
This may indicate a
decrease in the temperature necessary to convert anatase to rutile. which is also
consistent with H T X R D data. The H T X R D (in inert atmosphere) data showed that the
phase transition is 100°C lower than in air.
The increase in magnitude o f dielectric loss is likely indicative o f an increase in
electronic conductivity.
Conductivity (dielectric
loss) continues to increase as
temperature rises, indicating an increase in the defect concentration in the nitrogen
system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
200
400
600
800
1000
200
0
2C0
400
600
SCO
IOCO
1200
|400
Figure 4.5. Dielectric Measurements o f the Reaction between A l;0 , and T iO ; as
Temperature Increases in Air
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
:oo
400
600
soo
1000
1200
1400
1.2
453SiHT
/1
/
tan*
Oil
XO
400
oOO
300
IOOQ
1200
'.400
Fiaure 4.6. Dielectric Measurements ot'the Reaction between A l;0 - and T iO - as
Temperature Increases in Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
4.3.4 Aluminum titanate synthesis from defect rutile in nitrogen
Table 4.5. listing the X R D results o f the microwave synthesis o f A l , T i 0 5 from
mixtures o f alumina and defect rutile, shows that A l:T i 0 5 can be nucleated as low as
600°C. A t 750°C the sample reaction is about 60% completed. At 1250°C the reaction
is complete. As the product phase becomes the major phase the reaction rate is probably
limited by the diffusion processes through the product interfacial layer73,74 75. Since the
defect rutile-alumina mixture was dry-milled instead o f wet-milled prior to processing
to prevent the oxidation o f
the T iO :.x phase, the homogeneity o f the reactants is
questionable. This could lead to pockets o f alumina and titania that may not maintain
good intimate contact with the other reactant, decreasing the number o f reaction zones.
Table 4.5.
X R D Analysis o f the Microwave Synthesis o f A l:TiO< using A U O , and
Defect T iO , in Nitrogen
Conditions
Phases Identified by X R D and Relative Intensities
Temp/Soak
TiO, (rutile)
750UC/10 min
1050UC/15 min
1 1 5 0 T /1 5 min
1250UC/15 min
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
Dielectric measurements (Figure 4.7) made on the alumina-defect rutile system
show overall increase in &and e ' in comparison to the case o f alumina-anatase in
nitrogen synthesis. What is interesting to note is the continuous rise in the conductivity
(&') as the mixture is heated.
From the dielectric studies it is difficult to pinpoint the
nucleation temperature o f A l2T i 0 5. However, the regions where changes in the slopes
o f &and tii occur (starting at about 430°C followed by a brief plateau at 600°C ) might be
the nucleation point. This agrees with the X R D data. In the actual microwave chamber,
the reactants can be heated at about 60% o f the total power (about 0.78 K.W) without the
use o f secondary coupling.
Heating rates have been found to be near 200°C per minute
in some experiments after the temperature reaches about 400°C .
In the case o f the
alumina-anatase nitrogen synthesis. SiC rods were needed as secondary couplers and at
no point in the reaction did heating rates reach 200°C/ per minute.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
0
:oo
200
400
400
<00
<00
too
100
1000
IUOO
1200
1200
1400
1400
Figure 4.7. Dielectric Measurements o f the Reaction between A l- 0 ; and T iO ;„ as
Temperature Increases in Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
Figure 4.8, 4.9 and 4.10 are the plots of heating rates o f each sample against the
sample's bulk temperature for A l2T i 0 5 synthesized using T i 0 2.x, T i 0 2 heated in air and
T iO ; heated in nitrogen, respectively. Representation o f thermal data in this form may
allow for better understanding o f energy absorption trends and occurring phenomena. It
can be observed from Figure 4.8 that after the initial introduction o f microwave power,
the heating rate increased considerably as temperature increased until about 640°C . The
heating rate began to fall at that point, possibly indicating loss o f coupling ability due to
the disappearance o f T i0 2.x.
Figure 4.9 shows a rather linear heating rate, mostly
reflecting the heating o f the secondary coupler (SiC).
At no point in the reaction
process does the heating rate show a significant increase as seen in Figure 4.8.
4.10 does show a small increase in heating rate above 1000°C.
Figure
However, since these
plots were made from manual time-temperature measurements instead o f computerassisted or chart recorder measurements, the rise in heat rate may reflect a certain degree
o f error.
Figure 4.11 is the heating rate plot for T i 0 2.x alone.
The heating rate o f T iO ;.x
begins to fall at about 250°C, which is contrary to the reported data in the literature on
the heating o f T i 0 2.x72.
reflection is likely.
However, since T i0 2.x was microwave processed energy
As the temperature increases. T iO ;.x should become more
conductive, and more microwave absorptive to a point. As its conductivity increases it
w ill move closer to metallic behavior and begin to reflect microwave energy11.
Additionally it may also be possible for T i 0 2.x to reduce to a stoichiometric phase such
as T iO or T i20 3, which may have completely different microwave absorption properties.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
500
|
400
u
U-l
300
H
<
Od
O
z
200
H
uj
100
0
200
400
600
800
1000
TEMPERATURE (°C)
Figure 4.8. The Heating Rate Plot o f the Reaction between A l;0 3 and T iO :., heated in
Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
so
150
100
HHATING
RATH
(°<_7min)
200
0
200
400
600
800
1000
TEMPERATURE (°C)
Figure 4.9. The Heating Rate Plot o f the Reaction between A L O , and TiO - heated in A ir
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S!
R A TI: (°C7m in)
250
200
150
MHATING
100
0
200
400
600
800
10CO
1200
TEMPERATURE (°C)
Figure 4.10. The Healing Rate Plot o f the Reaction between A l;0 , and T : 0 : heated in
Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
s:
140
HEATING
RATH (°CVmin)
120
100
8 0
6 0
4 0
20
0
0
200
4 0 0
6 0 0
8 00
TEMPERATURE (°C)
F ig u r e
4 .1 1 . T h e
H e a tin g
R a te
P lo t o f T iO ;.4 h e a te d
in
N itr o g e n
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
In order to determine
the absorption
characteristics o f
T iO : .x, dielectric
measurements were carried out in nitrogen (Figure 4.12) and forming gas or reducing
atmosphere (Figure 4.13).
The absorption characteristics o f T iO ;.x under both
conditions are clearly different.
In nitrogen the dielectric measurements essentially
follow the heating rate plot for T i 0 2.„ having maximum peaks in & and e./ occurring at
about 600°C .
After 600°C both & and &' decline, as did the heating rate for T iO : .x in
Figure 4.11.
In a reducing atmosphere both & and
show continual increases.
This would
indicate that T iO :.x becomes more absorptive i f microwave processed in a reducing
atmosphere. Due to the strong microwave absorption o f T iO :.x heated in forming gas.
accurately measuring the absorption was not possible using the existing equipment.
Hence any microwave synthesis o f a material using T iO :.x as a reactant in forming gas
would likely result in further increases in reaction as well as diffusion rates.
X R D analysis o f the T iO : .x heated to 1500°C in forming gas during dielectric
measurements indicates rutile.
Hence reduction o f T iO ;.x to any stoichiometric T i-0
phase, such as TiO or T i20 3, did not occur.
However, the pattern could not be fully
identified because o f the existence o f smaller peaks surrounding the major peaks of
rutile. It is possible that the ion positions may have changed due to increased number of
vacancies after processing. This may have given rise to the excess peaks in the pattern.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.12 Dielectric Measurements ofTiCK., as Temperature Increases in Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 4.13 Dielectric Measurements o f T i O ^ as Temperature Increases in Forming Gas
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
86
4.3.5 Microstructural comparison o f defective and conventional A I2T i 0 5
Figure 4.14 and 4.15 are SEM micrographs o f a sample from the A l20 3- T i 0 2.<
reaction at 1250°C for 15 minutes and the conventional A l20 3- T i 0 2 reaction heated to
1500°C for 45 minutes, respectively.
Micrographs
and E D A X measurements were
made on the bulk o f the samples. Both microstructures show relatively uniform grain
sizes and a spiral-like pattern that possibly emanates from nucleation centers.
Hence,
similar microstructures can be obtained in microwave processing samples at nearly
300°C and 30 minutes less than under conventional conditions.
4.3.6 Reoxidation o f defective aluminum titanate
Figure 4.16 gives the dielectric data for the A l20 3- T i0 2., reaction after A l2T iO .
synthesis in nitrogen, followed by the introduction o f air into the system at 1<)0()°C in
order to determine i f the dielectric constant and tan 6 values would return to the values
obtained in the stoichiometric synthesis.
This would indicate that the defect A l2TiO<
could be reversibly transformed to stoichiometric A l2T i 0 5.
synthesis
tool
via
the
stoichiometric material.
non-stoichiometric
route
with
the
hence providing a fast
final
product
being
This information could be valuable for other microwave-
svnthesized materials by using defect components.
From the data presented in Figure 4.16 it was found that defect A l2T i 0 5 could be
reoxidized within 5 minutes at the final temperature.
constant
value
closely
approximated
the
value
The real part o f the dielectric
obtained
during
stoichiometric
processing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
87
Figure 4.14. S E M Micrograph o f the M icrowave Reaction between A l:0 , and T iO :„
heated at 1250°C for 15 minutes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ss
Figure 4.15. SEM Micrograph o f the Conventional Reaction between A1:0 ; and T iO M
heated in Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
89
<**■. 1
Figure 4.16. Dielectric Measurements o f the Microwave Reaction between A K O , and
T iO :.x in Nitrogen, then Under A ir Purge
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
90
Additionally, the tan 8 value gave the same result as the stoichiometric case. Hence the
resulting A I2T i 0 5 made from reoxidation appears similar to stoichiometric material.
4.4 Summary
A l2TiO< has been synthesized in a microwave furnace using a - A l20 3 and T iO ; in
air and nitrogen.
It was found that before any product phase was formed the anatase
conversion to rutile had occurred.
During this work no anatase phase was found
existing simultaneously with A I2T i 0 5, as determined by X R D analysis.
It was also found that A l2T i 0 5 could be microwave synthesized from anatase in a
nitrogen atmosphere at about 1050°C.
initiate at 1160°C.
In air. the microwave reaction was found to
This would tend to indicate that the microwave field is coupling
with the defects that are generated in rutile in the presence o f nitrogen.
This was
confirmed by the dielectric measurements.
But it was shown in this work that A l2T i 0 5can be most effectively synthesized in a
microwave field using a -A l20 3 and reduced rutile. The reaction starts as low as 600°C
after a soak time o f 15 minutes. In conventional furnaces the synthesis does not occur
until temperatures over 1300°C are achieved.
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Chapter 5
M IC R O W A V E S Y N T H E S IS OF B A R IU M T IT A N A T E
5.1 Introduction
5.1.1 General
Barium metatitanate (B a T i0 3), is one o f the several phases that exist in the BaOT iO ; system, which is shown in Figure 5.1. The phase diagram for B aO -TiO : system
was first studied by Rase and Roy83 in the 1950s and later updated by several workers84.
B a T i0 3 is the equimolar product o f
this system as shown in the equation below.
Though various uses have been found for other phases that exist in the B aO -TiO ;
system85, the tetragonal form o f B a T i0 3 remains the most technologically important
material in electroceramic industry fifty years after its debut.
BaO + T iO := B a T i0 3
Barium metatitanate can exist at room temperature in two structures: the tetragonal
distortion o f the ideal perovskite form and the high temperature hexagonal form86 87.
The name "perovskite'” is taken from the naturally occurring mineral. C a T i0 3. A series
o f compounds, including B a T i0 3, are isostructural with C a T i0 3 and hence all these
compounds are called perovskites. Though the name perovskite is used to describe
primarily cubic A *: B+4X 3 compounds, it is also used to describe phases having
distortions o f the cubic or "ideal” structure87 and with many compositions.
Figure 5.2
shows other existing polymorphs o f B a T i0 3 and their respective lattice constants as a
function o f temperature86. The high temperature or hexagonal form o f B a T i0 3 can be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
92
an1
000
j
e*M tone, ^ t,p ^ , | .
? L ^
Si
l c^« teno,»h^aT"!
*i:c
JO.
n * U.
:.j * ".c.
•cc
B«0
Figure 5.1. The Binary Phase Diagram o f the BaO-TiO; System
(Top: by Rase and Roy. bottom; by Kirby and Wechsier
(reference<l0)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
93
Dielectric
constant
Rhombohedral
Cubic
Orthorhombie
10.000
5,000
Curie
point
-150
-100
-6 0
0
♦60
♦
100
♦
150
Tempereture, *C
Figure 5.2. Phase Transitions o f BaTiO,
(reference “ )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
stabilized at room temperature by the incorporation o f metal impurities in the
structure88. Structural defects caused by oxygen vacancies or trivalent titanium ion can
also stabilize this phase87,88.
The B a T i0 3 tetragonal perovskite phase is ferroelectric and has been widely used
in several applications, while the hexagonal phase is non-ferroelectric and is o f
practically
no technological
importance.
B a T i0 3 was the
first perovskite-type
compound found to be ferroelectric in the 1940s89. and since then it has been the most
extensively studied ferroelectric material89,90.
The stable form o f B a T i0 3 above the
tetragonally distorted phase is the cubic form, which is not ferroelectric because it is
centrosymmetric and therefore cannot be poled. The tetragonal form which exists below
130°C. the orthorhombic form which exists below 5°C. and the rhombohedral form
which exists below -90°C. can all be electrically poled.
The electrical polarizabilitv is attributed to the TiO„ octahedra in the structure
(shown in Figure 5.3)89. Above the Curie point ( 130°C) the T i* 4 ion lies at the center o f
the structure giving rise to the cubic form: however, below 130°C the T i*4 ions occupy
off-center positions causing distortion in the cubic lattice which gives rise to the
tetragonal symmetry.
Hence the crystal structure changes from cubic to tetragonal
having c/a= 1.01. As the ferroelectric transition occurs (below 130°C). a spontaneous
polarization occurs along one o f the 6 edges o f the octahedron. The direction o f this
polarization can be switched by applying a high electric field o f around 1-2 kV/cm.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0<
Figure 5.3. Electrical Polarization o f TiO„ Octahedra
(reference ,6)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
%
5.1.3 Applications
Barium titanate is the basis o f the world’s capacitor and electroceramics industry.
It has been in use as a dielectric material since the early 1940s86. It has been developed
recently, along with other perovskite materials, as a multilayer capacitor material (a
schematic is shown in Figure 5.4).
In multicapacitor system, large capacitances are
observed by stacking several 20-50pm thin ceramic foils89.
During processing each
layer contains screened noble-metal electrodes, mainly palladium, while in the green or
unsintered state. These layers are then laminated, pressed and sintered in air at I300°C .
The capacitance o f such capacitors is in the 100 n F -lp F range.
Other types o f
capacitors that use B a T i0 3 are disk capacitors, tube capacitors and multichip capacitors.
By doping B a T i0 3 with La on the A-site. or with Sb or Nb on the B-site. BaTiO-,
can be transformed into a semiconducting material. N-type semiconductor B aTiO } has
been used in positive temperature coefficient (PTC ) materials89. PTC materials are used
mostly in controlling or measuring applications such as current limiters, self-controlling
heating elements and deguassing units in television sets.
resistivity (10 Q-cm) at room temperature.
PTC ceramics exhibit low
As the temperature increases and
approaches the Curie point, the resistivity increases forming a resistor material.
PTC behavior originates at the grain boundaries where defects with acceptor
character trap conduction electrons. This forms a depletion layer with a high resistance
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
97
' Solderlandi
Electrode*
Ceramic material
End terminal
Figure 5.4. Schematic o f a M ulti-layer Capacitor
(reference ” )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
potential barrier.
The magnitude o f this barrier depends upon the ferroelectric
properties o f B aTiO ,.
5.1.4 Synthesis
Barium
titanate (B a T i0 3) is made
from several synthetic
routes, including
hydrothermal techniques91,92,93, sol-gel and coprecipitation techniques93,94, as well as
ordinary solid state reaction methods95,96,97,98. The synthesis method most commonly
used industrially is the solid state reaction route99,100 between B a C 0 3 and T iO : as
described by the equation below:
B a C 0 3 + T iO : = B a T i0 3 + C 0 3
Stoichiometric barium titanate has a dielectric constant in the range o f 1000-2Q0Q86.
depending on grain size and density.
Hvdrothermally synthesized B a T i0 3 has been produced by Hennings et al. from
titanium acetate sols and barium acetate in acetic acid. Gelation o f this mixture at room
temperature was slow, requiring up to 24 hours for solidification.
Increasing the
temperature to 60°C reduced gelation time to a few minutes93. The gel was then dried at
80-150°C .
The pH o f the gel was adjusted with tetramethylammonium hydroxide to
basic (p H > 13 ). This was necessary to maintain the thermodynamic stability o f aqueous
suspensions o f B a T i0 3 and prevent decomposition93. Tetramethylammonium hydroxide
was used as a 25% aqueous solution.
The hydrothermal reaction was completed at
150°C for 10 hours at a pressure o f 150 bar. The resulting product consisted o f 200300nm spherical particles.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Schmutzler et a l100 attempted the synthesis o f B a T i0 3 by the oxidation o f metallic
precursors. They used an equimolar ratio o f barium flakes with titanium powder. The
powder mixture was placed in a glass vial and milled in argon gas for several hours.
The mixture was then either m illed under cryogenic conditions in a rod m ill or at room
temperature in a high energy ball m ill for 1 to 3 hours. The powder mixture was loaded
into one-end closed silver tubes which were sealed and rolled into tapes.
Each rolled,
silver-sheathed tape had a dense B a-Ti core. The silver-sheathed precursor tapes were
cut and divided for processing. Samples were heated at several heating rates to several
temperatures up to 900°C, above the melting point o f the silver. To avoid formation o f
Ba-Ag liquid. Ba was preoxidized at 300°C
for 24 hours.
The samples were
characterized by XR D . Single-phase B a T i0 3 was obtained after heating above 900°C.
Solid state synthesis is the method by which most commercial BaTiO-, is produced.
Normally, B a C 0 3 and T i 0 2 are used as precursors95,96'97 9*. The reaction begins with the
decomposition o f B a C 0 3 to give BaO which then reacts with T iO : to give B a T i0 3.
Solid state B a T i0 3 synthesis is described in detail in section 5.3.2.
5.2
Experimental Procedures
5.2.1 B a C 0 3-T iO : system
Equimolar amounts o f B a C 0 3 (Mallinckrodt, Inc) and T iO ; (rutile) (Cerac. Inc),
having particle sizes o f 2pm and 5pm respectively, were dry-milled for 16 hours using
A1:0 3 media. The container used was a Nalgene™ bottle with a capacity o f 1 liter. The
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
charge roughly filled 90% o f the container. The milling speed was about 65 revolutions
per minute.
The media was separated from the reactant mixture by passing through a 40-mosh
screen with vigorous shaking.
The resulting powder was pressed into pellets with a
stainless steel die o f 2.873cm in diameter.
The pellets weighed about 12 grams and
were about 0.762cm thick. Powder compacting was employed in this study to increase
particle contact, and not for sintering studies. The pellets were pressed uniaxially using
9000-10,000psi. o f force in a Carver Lab press.
This technique was used for the
fabrication o f all test samples.
5.2.2 B a C 0 3- T i 0 2., system
TiCh. anatase phase. (Cerac. Inc) was reduced at U 0 0 °C for 3 hours by the
method described in section 4.2.2. Equimolar amounts o f B a C 0 3 (Mallincrodt. Inc) and
reduced T iO : . having particle sizes o f 2pm and 5pm respectively, were dry-milled and
pressed into pellets as in section 5.2.1.
5.2.3 Conventional synthesis o f Barium Titanate
5.2.3.1 Using B aC 0 3- T i 0 2.x
Each pellet was placed on an alumina crucible without the use o f a powder bed.
The pellets were heated in a tube furnace (Lindberg) in a nitrogen atmosphere, using an
Omega controller. Firebrick was used as thermal insulation for the end caps in the tube
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
101
furnace. Prior to the start o f each experiment, the furnace was purged with the process
gas for 30 minutes.
Samples were fired isothermally at 900°C, 950°C, 1000°C. 1100°C. 1200°C and
1300°C for 15 minute intervals from 15 minutes to 1 hour. As described in the previous
chapter, 4 samples were fired at a each temperature for different time periods.
The
heating rate was 20°C/min in all experiments. Each pellet was allowed to cool as the
furnace cooled.
information
Phase information was obtained using X R D .
was obtained
on
selected
samples by SE M
and
Microstructural
EDAX
analyses.
Additionally, the reaction was followed by a combination T G A -G C -1R method to
determine the evolution o f gases from the decomposition o f the reactants.
5.2.4 M icrow ave synthesis o f barium titanate
5.2.4.1 Using B aC 0 3- T i 0 2 powders
Each reactant pellet was placed in cylindrical alumina insulation and then
surrounded by A l20 3 beads.
board (Z A L -4 5 , Zircar).
The top and bottom o f the cylinder was alumina fiber
The entire assembly was placed inside the model 1300
atmosphere chamber as described in detail in chapter 3. Three SiC rods were used as
secondary couplers for air atmosphere or stoichiometric experiments. Before the start o f
each experiment, the atmosphere chamber was evacuated and backfilled with the
desired atmosphere 3 times to ensure proper atmosphere control.
For stoichiometric experiments 80% o f the total power was applied to initiate each
experiment due to the inability o f T i 0 2 and B a C 0 3 to couple at low temperatures. For
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
stoichiometric experiments, the heating rate was 20°C/min starting at 50°C below each
set point temperature to minimize the possibility o f overshooting the set point
temperature.
The
microwave
experiments
were
analogous
to
the
A l:TiO<
microwave
experiments with respect to the soak times used at the various temperatures. However,
the temperatures used in these experiments were 900°C, 950°C, 1000°C and 1050°C.
Each sample was allowed to cool naturally without quenching.
Each sample was
characterized by X R D after processing for phase identification, and the microstructures
o f selected samples were examined by SEM .
5 .2 A 2 Using B a C 0 3-T iO :.x powders
Each pellet was placed in alumina housing surrounded by A1i 0 3 beads inside the
model 1300 atmosphere chamber o f the microwave unit as described in detail in
Chapter 3. No secondary coupler was necessary for the defect-TiCK experiments. The
vacuum and backfilling procedure was the same as described previously in section
4.2.5.
Only 60% o f the total power was required, since the samples coupled
almost
immediately. Temperature control was difficult because o f thermal runaway condition:
hence it was necessary to lower the power prior to soak temperature.
The maxim um soak time used in these experiments was 2 minutes.
temperatures attempted were 400°C, 500cC, 600°C. 700°C, 800°C and 900°C.
The
These
samples were also allowed to cool without quenching. Phase information was obtained
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
103
on each sample by X R D and microstructural analysis was carried out by SEM and
EDAX.
5.3 Results and discussion
5.3.1 T G A -G C -IR analysis o f the conventional B a C 0 3-TiCh reaction
Samples o f the B aC 0 3-T iO : reaction mixture were analyzed in air and nitrogen
atmospheres by a combination of T G A -G C -IR methods. This was done to determine
the temperature for the onset o f B a C 0 3 decomposition and to determine the volatile
components as the reaction progresses.
Figure 5.5, the T G A data obtained as the
temperature increases, shows very little weight loss from room temperature to about
850°C in either atmosphere. 1R analysis showed that the weight loss was primarily due
to water evaporation with small amounts o f C O ; gas.
enhanced and by 1370°C
Above 850°C weight loss was
the sample lost nearly 10% o f its original weight, the
maxim um amount possible. IR analysis showed that this weight loss was due to C O ;
evolution only.
5.3.2 Conventional B a T i0 3 synthesis by solid state reaction using T iO i.x
Conventional
documented.
work.
synthesis o f B a T i0 3 using
stoichiometric
precursors
is well
Hence conventional stoichiometric synthesis was not attempted in this
However, it is important to compare the results o f conventional non
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
ioH
io<H
90H
56- i
32-
90 H
200
400
600
600
1000
1400
Teaoerature C
Figure 5.5. T G A Thermogram o f the Reaction between BaC O :. and T iO ,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
-stoichiometric synthesis with microwave non-stoichiometric synthesis. The reactants
used for conventional non-stoichiometric synthesis were also used for microwave nonstoichiometric
synthesis.
Using reduced
or non-stoichiometric
components
for
conventional processing is likely not a common practice and no literature was found on
that topic.
The phase diagram for the B a O -T i0 2 system (Figure 5.1), which was initially
determined by Rase and Roy and later updated by Kirby and Wechsler. can be followed
to explain the reactions observed from conventional non-stoichiometric synthesis. The
X R D results in Table 5.1 show the steps in the formation o f B a T i0 3 synthesis.
Reaction begins above 900°C as described by Beauger95 and other workers9* 97 9S
with the formation o f Ba2T i0 4. The reaction continues until Ba2T i 0 4 is the only phase
detectable by X R D analysis after 1 hour at 1300°C.
Since the B aC O :, -T iO ; .x mixture
was equimolar. the presence o f a T iO :-rich phase is expected, however, none was
detectable by X R D . This contradicts the reaction pathway described by Beauger et al9h.
The reaction pathway for B a T i0 3 synthesis described by Beauger proceeds by the
sequence shown in the following equations:
B a C 0 3 + T i 0 2= B a T i0 3 + CO
(A )
B a T i0 3 + B a C 0 3= Ba2T i0 4 + C O :
(B)
Ba2T i0 4 + T iO ; =2BaTiO
(C)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
106
Table 5.1. X R D intensities from the conventional BaCCb-TiCb-x synthesis
Condition
_________ Phases Identified by X R D and Peak Intensities
Tem p/ Soak
BaCCb
TiCb.x
9 0 0 °C /2 min
100
75
9 5 0 °C /2 min
100
50
950°C/1 hour
100
37
23
U 0 0 °C /1 hour
94
100
65
1200°C/1 hour
68
85
77
BaTiCb
Ba2TiCb
100
1 3 0 0 ° C /1 hour
They claim that first a small amount o f BaTiCb is formed from the reactants at the
expense o f TiCb, which is reaction A. This reaction begins near 750°C %. By 900°C it
was reported that a homogenous layer forms consisting o f a mixture o f BaTiCb and
BazTiO.*.
BaiTiCb results from a reaction with BaCCb and BaTiCb.
This reaction
occurs until the BaCCb completely disappears, after which Ba 2TiCb reacts with the
remaining TiCb to give BaTiCb as the final product.
The results shown in Table 5.1 suggest a different reaction pathway.
Since no
TiCb was observed by X R D analysis and amorphous TiCb phase exists at high
temperatures, it is probable that an amorphous Ba-containing phase rich in TiCb has also
been formed at this temperature (1300°C ).
As temperature increases this amorphous
phase may react with Ba2TiCb to give BaTiCb. This mechanism is essential because the
starting mixture was equimolar.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
107
5.3.3 Stoichiometric microwave synthesis
The X R D results from the reaction between stoichiometric T iO : and B a C 0 3 are
given in Table 5.2.
The table shows the progression o f the reaction from 900°C to
1050°C under various soak times. As can be observed, the reaction follows the same
pathway as in the conventional case, beginning with the formation o f Ba2T i0 4.
However, there is a radical distinction by 950°C : B a T i0 3 begins to form, and. as the
temperature rises, it becomes one o f the major phases along with Ba2T i 0 4.
In comparing the conventional case with the stoichiometric microwave case, it can
be readily observed that conventional heating did not produce B a T i0 3 below 1300°C. as
was shown in Table 5.1.
Additionally. Ba2T i 0 4 was the only barium containing
crystalline phase after 1 hour at 1300°C, whereas the microwave case had essentially
equal concentrations o f B a T i0 3 and Ba2T i 0 4 as low as 1050°C. However, since BaTiO-,
is present in increasing amounts with increasing temperature, a competition between the
reaction to form Ba2T i0 4 (from B aC 0 3 and B a T i0 3) and the reaction o f Ba2T i 0 4 with
the amorphous T iO ;-rich Ba phase to form B aT iO j is possible. This would indicate that
microwave processing may enhance the reaction o f Ba2T i 0 4 with the amorphous T iO :rich Ba phase to form B a T i0 3 as opposed to the conventional case. Hence, the
influence o f the microwave field is clearly beyond mere rapid heating.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
108
Table 5.2. X R D intensities from the microwave BaCC^-TiOi reaction
Condition
Phases Identified by X R 3 and Peak Intensities
Temp/Soak
BaCCb
T iO i
900°C /15 min
100
99
9
900°C/1 hour
98
100
30
950°C /30 min
100
88
18
25
950°C/1 hour
82
100
20
30
1000°C/30 min
67
82
45
100
1050°C/30 min
72
71
83
100
BaTiCb
Ba2TiCb
5.3.4 M icrowave synthesis o f BaTiCb from non-stoichiometric T iO j
A strikingly new nucieation and reaction pathway is demonstrated in this case. The
X R D results are shown in Table 5.3 for the reaction between BaCCh and nonstoichiometric T i 0 2 (or TiOi-x).
It can be observed that near single phase BaTiCb has
been completely synthesized by 700°C (bulk temperature) with only trace amounts o f
the starting reactants left. A t a bulk temperature o f only 250°C, a hexagonal BaTiCb
phase has already been nucleated and by 400°C it is the major phase.
Tetragonal
BaTi 0 3 is also present at 400°C along with amounts o f the starting reactants. A t 600°C
the tetragonal phase is the major phase, while the hexagonal phase is disappearing. At
900°C the reaction is nearly complete containing tetragonal BaTiC^ and trace amounts
o f T i 0 2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
109
Table 5.3. X R D intensities from the microwave B aC O j-T iO j., reaction
Condition
Phases Identified by X R 3 and Peak Intensities
Temp/Soak
B aC 03
T i0 2
H ex-B aTiO j
250°C /0 min
87
100
7
4 00 °C /0 min
52
45
79
12
500°C/1 min
22
18
100
44
6 00°C /5 min
24
28
63
100
700°C /5 min
9
12
23
100
900°C /5 min
2
Tet-B aTiO j
—
100
It is also striking that the reaction occurs without the formation o f any Ba2Ti0.t. Since
the hexagonal phase is stabilized first, a completely different reaction pathway seems to
be operating.
Phase diagrams for BaTiCb show the hexagonal phase is stable at
temperatures over 1400°C (Figure 5.1); however, the hexagonal phase can also be
stabilized at lower temperatures by vacancies and non-stoichiometric defects87,88. Since
the process was carried out in an inert atmosphere, starting with a defective reactant, and
heating was rapid, the nucleation o f the metastable hexagonal phase seems plausible.
The formation o f the metastable hexagonal phase at these low temperatures may be
due to the fast heating rate. Schmutzler et al. reported hexagonal BaTiCb formation in
the metallic oxidation synthesis100. The authors attributed nucleation o f the hexagonal
phase to the formation o f molten BaO at 450°C when the reactant mixture was heated
rapidly from 300°C to 500°C. They suggest that the exothermic T i reaction with BaO to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
110
hexagonal phase formation. This m ay not explain the phenomenon entirely, because the
heating rate reported was only 25°C /m in.
However, when the heating rate was low
(l° C /m in ) no hexagonal phase was formed, only Ba:T i 0 4.
The subsequent formation and continual growth o f the tetragonal phase, as well as
the disappearance o f the hexagonal phase, as the temperature increases is probably
thermodynamically driven.
This is quite likely because, from the phase diagram, the
tetragonal phase is thermodynamically stable in this temperature range.
5.3.5 Heating effects in non-stoichiometric synthesis
In these experiments what was probably occurring was a rapid reaction between
T iO :.x and BaO.
The cause o f this diffusion was no doubt the "superheating" o f the
T iO : .x as explained in chapter 4. even though the bulk temperatures were low (<400°C).
The temperature o f the T iO :.x particles may have been much higher. Figures 5.6 and 5.7
are plots o f heating rate vs. temperature for the reactions o f BaCO: with T iO :.x and
T iO ; . In both experiments the microwave power was kept constant to eliminate effects
o f increasing or decreasing power so that increases or decreases in heating rates would
be functions o f the sample only. In the early stages o f reaction the heating rate o f both
systems showed gradual rise.
A t about 170°C, the T iO ;.x reaction began to undergo
thermal runaway and by 500°C had a heating rate o f about 480°C/m in. whereas the T iO ;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
500
HEATING
RATH (°C/min)
4 0 0
3 0 0
200
100
0
200
4 0 0
6 0 0
8 0 0
1000
TEMPERATURE (°C)
Figure 5.6. The Heating Rate Plot o f the Reaction between BaCO, and T iO ;.v heated in
Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HEATING
RATO (°C/m in)
112
250
200
150
100
0
200
400
600
800
1000
1200
TEMPERATURE (°C)
Figure 5.7. The Heating Rate Plot o f the Reaction between BaCO-. and T iO , heated in
Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
113
reaction maintained a relatively linear heat rate at all times. This was likely the start o f
the reaction. Due to the fast rise in the temperature, it was not possible to examine the
sample for content during the initial stages.
Above 500°C , the heating rate began to
gradually decrease. Though the reaction was not complete, it may indicate that the
reaction mixture was becoming less energy absorbing due to reaction and consequent
loss o f the most absorptive phase, T iO : .v
5.3.6 Microstructural examination o f B a T i0 3 specimens
SEM
micrographs were taken o f the B a C 0 3- T i 0 2.x reaction
mixture after
processing at 1000°C for 15 minutes ( Figure 5.8). These experiments were attempts to
sinter already formed B a T i0 3.
The results show likely melting o f the material with
large grains forming, possibly due to the superheating o f T iO :.v
In the micrographs
and E D A X analyses, it can be seen that melting has resulted due to a lamellar
microstructure that included an aluminosilicate impurity phase from the alumina beads
used as thermal insulation.
boundaries.
Additionally. Ba6-Ti-17 phase appears around grain
This phase has been seen surrounding grain boundaries o f B a T i0 3 melts
after crystallization.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
114
1 00R m
526
15KU
X 100
12mm
Figure 5.8. S E M Micrograph o f the Reaction M ixture from BaCO, and T iO :., heated at
1000°C in Nitrogen for 15 Minutes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
5.4 Summary
The above data suggest that a "microwave effect" exists for the synthesis o f BaTiO,
in a microwave field. This effect is manifested by the enhancements in. or departure
from, conventional reaction pathways. The microwave synthesis attempted in air shows
that B a T i0 3 is present in considerable amounts along with Ba:T i 0 4, which does not
happen in conventional synthesis.
B aT iO j synthesis using T iO : .x has shown some different results vs. other solid
state methods. The rationale is likely the ability o f defective T iO ; to couple effectively
to the applied energy and "superheat” in a thermal runaway mode. This situation greatly
enhances the reactivity of the T iO ; through the reaction medium, as was noted in the
case o f A I; T i 0 5 (chapter 4). causing the reaction to proceed via different routes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I 16
Chapter 6
M IC R O W A V E S Y N TH E S IS OF B A R IU M M A G N E S IU M T A N T A L A T E (B M T )
6.1 Introduction
6.1.1 General
Barium magnesium tantalate (B M T ) (BajM gTaiOg or B a f M g i ^ T a ^ ^ ) 101102 has
the perovskite structure. It is one o f several perovskite structured materials that are
being used as microwave substrate materials101 102 103,104105 10610710*.
B M T prepared
from a solid state reaction has been found to have very low sinterability unless M ndoping is employed or rapid tiring at accelerated heating rates103. The low sinterability
can be explained by the B M T melting point. B M T has been shown by Guo and Bhalla
in this lab to melt congruently at about 3000°C. which makes it possibly the highest
melting oxide known.
This is an extraordinary observation for such a complex
composition and one with potentially high anisodesmicity with Ta5* and M g: ‘ in
equivalent octahedral sites. Hence, this is an interesting and technologically important
material for processing by a new method.
6.1.2 Applications
B M T is an excellent candidate for microwave substrate purposes as well as a
candidate for high temperature superconductor (H TS C ) substrates101 because o f its high
Q or very low dielectric loss. The Q-value can be represented as the inverse o f the loss
tangent, tan S. Using 1 mol% M n as a sintering aid, B M T was reported to have a Q-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
117
value o f 16800 at 10.5 G H z and a dielectric constant o f 2 5 l02. This is the highest Qvalue reported for an oxide material. However, B M T has been found to have varying Qvalues, possibly due to variations in the ordering in the B site o f the perovskite phase102
and in the sintered density. The dielectric constant is unaffected by B site orderingIn:.
Because, B M T has excellent microwave dielectric properties, finding simpler ways to
synthesize and sinter this material could be very valuable.
6.1.3 Synthesis techniques
B M T has been synthesized by several techniques, including laser102 and a metal
alkoxide hydrolysis route101. Guo. Bhalla and Cross reported growth o f single crystal
B M T using the laser-heated pedestal growth technique (L H P G )102.
Katayama and
Sekine used a metal alkoxide hydrolysis route involving tantalum ethoxide. Ba and Mg
metal as starting materials. The molar ratio was 3:1:2 (Ba:Mg:Ta).
Ba and M g metal
were added to a solution o f tantalum ethoxide and ethanol. This mixture was refluxed
for 12 hours in dry nitrogen, resulting in a completely homogeneous solution o f Ba. Mg
and Ta alkoxides.
Water was added to the solution followed by further refluxing or
stirring for 15 hours at room temperature. The resulting white precipitate was separated
from solution by ultrafiltration and dried at 110°C for 48 hours. This precipitate was
then calcined for 3 hours. X R D showed a perovskite phase at a calcining temperature o f
400°C , though the sample was not fully crystalline.
The authors report sintered
densities o f nearly 98% o f theoretical density for B M T sintered at 1400°C for 8 hours,
preceded by calcining at 1000°C.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
118
Solid state synthesis o f B M T . such a high melting oxide, requires very high
temperatures. Guo, Bhalla and Cross102 synthesized B M T by mixing high purity M gO .
Ta20 ; and B aC O j and heating in an alumina crucible at I500°C for 3 hours. Sintering
o f B M T made from this route required 3-35 hours above 1600°C.
6.2 Experimental Procedures
6.2.1 Synthesis o f stoichiometric B M T from B a(O H )i. M gO and Ta20 5
Stoichiometric B M T was microwave synthesized following only one route in which
B a(O H )2.M g O and Ta20 5 were used as precursors. The results o f this synthesis were
compared with those o f the non-stoichiometric B M T syntheses. The heating rate plots
o f the two B M T synthesis techniques were compared and will be discussed in section
o .j .j .
Chemically stoichiometric amounts o f B a(O H )2 (Cerac). M gO (Fisher)and T a :0<
were first dry m illed using A l20 3 m illing media inside o f a Nalgene rM bottle having a
capacity o f 1 liter.
Dry milling was done, as opposed to wet milling, to prevent the
hydrolysis o f BaO to give Ba(O H )2.
The resulting powder was separated from the
m illing media by pressing it through a 40 mesh screen. Pellets were uniaxiallv pressed
using a Carver Lab press with no binder material.
Pressed pellets had diameters o f
2.873cm. weights o f about 12 grams and pellet heights o f about 0.762cm.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
II*)
6.2.2 Reduction o f tantalum pentoxide
Reduction o f Ta20 5 was carried out using a high temperature furnace (Astro, Inc).
Samples were heated in boron nitride crucibles at 1450°C for 2 hours in a 5%hvdrogen95% argon atmosphere (Linde Inc.)79.
Phase information was obtained by X R D
analysis.
6.2.3 Synthesis o f B M T using non-stoichiometric Ta:0 5
Three m ixed oxide formulations were attempted for non-stoichiometric B M T
synthesis, using defect Ta20 5 (Ta20 5.x) in all non-stoichiometric experiments.
The
details are presented below:
A. B a C 0 3-M g 0 -T a 20 5.x system
Chemically stoichiometric amounts o f B a C 0 3, M g O and reduced T a 20< were dry
milled using A120 3 media in a Nalgene™ container.
Dry milling was employed to
prevent the oxidation o f the Ta20 5.x phase. The ball-m illing speed was 65 rpm. After
milling, the powders were separated from the media via vigorous sieving using a 40mesh screen.
In order to increase the particle contact area, the powders were pressed into pellets
using a Carver Lab press at a pressure o f 10,000 psi. The powders were pressed in a
stainless steel die having a diameter o f 2.873cm, The weights o f the pellets were about
12 grams and the pellet heights were about 1.016cm.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
B. B a(O H )i x H 20 - M g 0 - T a 20 5.x system
Chemically stoichiometric amounts o f B a(O H )2 x H 20 , M gO and reduced Ta20<
were dry m illed using A120 3 media in a Nalgene™ container.
Dry m illing was
employed to prevent the oxidation o f the Ta20 5.x phase. The ball-milling speed was 65
rpm.
A fter m illing, the powders were separated from the media via vigorous sieving
using a 40-mesh screen.
The resulting powders were pressed into pellets having diameters o f 2.873cm,
weights o f about 11 grams, and pellet heights o f about 1.016cm.
C. B a 0 -M g 0 -T a 20 5 system
Amounts o f BaO, M gO and reduced Ta20 5, according to equation 1, were drymilled using A l20 3 media in a Nalgene™ container.
Dry milling was employed to
prevent the oxidation o f the Ta20 5.x phase. The ball-m illing speed was about 65 rpm.
After milling, the powders were separated from the media via vigorous sieving using a
40-mesh screen.
The resulting powders were pressed into pellets having diameters o f 2.873cm.
weights o f about 12 grams, and pellet heights o f about 0.762cm.
6.2.4 M icrowave synthesis o f B M T from stoichiometric Ta20 5 routes
Each pellet was placed in alumina housing surrounded by hollow bubble A l20 3
insulation beads inside the M M T model 1300 atmosphere chamber o f the microwave
unit as described in detail in Chapter 3. These experiments required the use o f 3 SiC
rods as secondary couplers.
Before the start o f each experiment the atmosphere
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
chamber was evacuated and backfilled with the desired atmosphere 2-3 limes to ensure
proper atmosphere control.
For stoichiometric experiments 80% o f the total power was applied to initiate each
experiment due to the inability o f the starting reactants to couple at low temperatures. In
the case o f the stoichiometric experiments, the heating rate was 20°C /m in starting at
50°C below each set point temperature to m inim ize the possibility o f overshooting the
set point temperature.
The temperatures used in these experiments were in the range o f 71 °C -1100°C .
Each sample was allowed to cool without quenching. A fter each experiment the samples
were characterized by X R D for phase composition.
6.2.5 Microwave synthesis o f B M T from non-stoichiometric Ta;0 ,
Each sample pellet was placed in alumina housing surrounded by hollow bubble
A U 0 3 insulation beads inside the M M T model
microwave unit.
1300 atmosphere chamber o f the
No secondary coupler was used for certain Ta20 5.x experiments.
It
was found that 3 SiC rods had to be used as secondary couplers. The evacuation and
backfilling procedure used was the same as in section 6.2.4.
These experiments were
all conducted in nitrogen atmosphere.
Ta20 5.x experiments required about 80% o f the total power, since coupling o f the
material was not as effective as in the case o f T i 0 2.x.
analogous to those used in section 6.2.4
The temperatures used were
The samples were allowed to cool without
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
quenching.
A fter each experiment the samples were characterized by X R D for phase
composition.
6.3 Results and Discussion
6.3.1 Synthesis o f reduced Ta20 5 powders
Reduced/defect Ta20 5 was synthesized at 1450°C for 2 hours.
Figures 6.1a and
6.1b display X R D patterns for Ta20 5 and the reduced material, which is designated as
Ta20 5.x.
The X R D patterns indicate that Ta20< has an orthorombic structure, and that Ta20 5.x is
tetragonal, which is consistent with the information reported in the literature74.
The
conductivity studies carried out on Ta20 5 and Ta20 5.x give conductivity values o f
3.4x1 O'8 (D -cm )"1 and 2.3x1 O'6 (D - c m f 1 respectively, in agreement with literature
results74. Hence, as in the case o f T i 0 2.x, Ta20 5 can also be made more conductive by
introducing oxygen vacancies in the lattice, but the resistivity o f Ta20 5.x is orders o f
magnitude greater than that o f T i0 2.x.
6.3.2 M icrowave synthesis o f non-stoichiometric B M T using B aC 03
The X R D results from the non-stoichiometric B M T synthesis using B a C 0 3 as the
barium source are shown in Table 6.1. By this route. B M T could not be formed up to
1100°C for 15 minutes.
In the experiments represented by data in Figure 6.2. no SiC
was used as a secondary coupler.
The microwave conditions were similar to the
conditions reported for syntheses using T i 0 2.x, since Ta20 ;.x was also found to be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
123
i
laasa
aaae
zaaa
2
a
38
ia
sa
ta
sa
60
(count*I
saaa
teaa
zaaa
zaaa ■
taaa
za
Figure 6.1. X R D Patterns for (a) Ta;0 , and(b) T a ;0 ;.x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
124
conductive.
However, the bulk temperatures in any o f these experiments never
exceeded 150°C and had extremely low heating rates as shown by Figure 6.2.
When
SiC was employed for secondary coupling a bulk temperature o f 1100°C was achieved
and no evidence o f thermal runaway or high heating rates was observed. Also in that
experiment no B M T phase was detected by X R D analysis. In this particular system no
microwave enhancements were noted up to 1100°C.
Due to the poor coupling
conditions and no evidence o f increasing heat rates, temperatures above 1100°C were
not investigated.
Table 6.1. X R D peak intensities from the BaCCVM gO-TaiCh.* synthesis
Condition
X R D Phase Identification and Relative Peak Intensities
Tem p/Tim e
raiO<.x
77"C70 min
B aC O j
MgO
'45
lOu
37
127uC /0 min
53
\M
48
138°C /0 min
39
100
30
1100uC/15m
6 0 - L . 100-a
70
38
"
■BMT
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 :5
HEATING
RATE (°C/min)
12
10
6
6
4
20
40
60
80
100
120
140
TEMPERATURE (°C)
Figure 6.2. The Heating Rate Plot o f the Reaction between B a C 0 3. M g O and Ta-0< v
heated in Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
126
6.3.3 M icrowave synthesis o f non-stoichiometric B M T using a B a(O H b route
Because B M T could not be synthesized using BaCOj, B a(O H ): was attempted.
The X R D results are given in Table 6.2, and each experiment underwent the same
conditions as the B a C 0 3 route using no SiC. From these results it can be seen that B M T
was synthesized from this route as low as 253°C. By 532°C. B M T was the major phase.
However, the reaction was not completed by 781°C.
reactants remained even after heating at 781°C.
Considerable amounts o f the
It was not possible to achieve higher
bulk temperatures without the use o f SiC in the microwave system.
Figure 6.3 is the
heating rate plot for the 766°C reaction. A t 766°C. as at 78 l°C . the bulk temperature
fell even though the microwave ftimace was operating at full power. This indicates that
the sample ceased to couple with the applied field.
Table 6.2. X R D peak intensities from the Ba(0H)->-Mg0-Ta->0<.x reaction (without use
o f SiC)
Condition
X R D Phase Identification and Relative Peak Intensities
Temp/Soak
B a (O H ),
....... M g O "
■BMT""
532uC /0 min
32
20
8
100
569uC /0 min
44
30
3
100
766“C /0 min
40
10
10
100
781 “C /0 min
42
14
10
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
100
HEATING
RATH (°C/min)
120
0
50
100
150
200
250
TEMPERATURE (°C)
Figure 6.3. The Heating Rate Plot o f the Reaction between B a(O H ); . M g O and Ta-Os*
at 766°C in Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
128
There are two possible explanations for the loss o f coupling ability by the sample.
The first may be oxidation o f Ta20 5.x due to water loss of B a (0 H )2x H 20 , and then its
decomposition into BaO and H 20 . Obviously B a(O H )2 was the initial phase which was
coupling with the microwave field at low temperatures due to water emanation. This is
demonstrated by the equation below:
B a (0 H )2x H 20 = BaO + ( l+ x ) H 20
(1)
Hence, it is possible to oxidize Ta20 5.x in the presence o f B a (0 H )2x H 20 .
Once
Ta20 5.x is oxidized or approaches stoichiometry, the result should approximate the
microwave synthesis in air. The second possibility is the fact that B M T . which is the
major phase at the point o f coupling loss, does not couple with microwave energy due to
its high Q value (discussed earlier).
In order to determine which mechanism is operating here, microwave synthesis of
B M T using stoichiometric Ta20 5 in air and using Ta20 5.x in nitrogen were conducted.
Figures 6.4 and 6.5 are the heating rate plots o f the microwave synthesis using Ta20 5 (in
air) and Ta20 5.x (in nitrogen) respectively. Both syntheses were aided by use o f SiC as a
secondary coupler.
The air synthesis was carried out up to 1400°C, and the nitrogen
reaction up to 1200°C. X R D analysis showed that in both cases the reaction was nearly
complete, with only trace amounts o f unreacted starting materials.
The Ta20 5.x sample in nitrogen shows two regions of maximum microwave
coupling peaking around 290°C and 1020°C. respectively. The peak at 290°C likely
represents the dehydration and decomposition o f Ba(O H )2 to BaO, and the peak at
1020°C likely represents the completion o f B M T synthesis.
This indicates that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HEATING
RATE (°C/min)
80
60
40
20
0
0
200
400
600
800
1000
1200
1400
TEMPERATURE (°C)
Figure 6.4. The Heating Rate Plot o f the Reaction between B a(O H ): , M gO and Ta:0 ,
heated in Air
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
130
200
S
E
150
U
0
til
h
<
100
Q£
Z
r-
<
U
I
0
200
400
600
800
1000
1200
TEMPERATURE (°C)
Figure 6.5. The Heating Rate Plot o f the Reaction between Ba(O H ): . M g O and Ta;0 ,„
heated in Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
131
mixture w ill couple after raising the temperature beyond where the sample ceases to
couple initially.
Since the initial rise o f the I020°C peak is at about 720°C . it is not
likely that water oxidizes Ta;0 5.x.
However, it is possible that initial coupling loss and subsequent decrease in heating
rate above 290°C are due to B M T formation. Since the B M T phase formed is probably
o f a defective non-stoichiometric nature due to Ta:0 5.x. use o f a secondary coupler may
increase the bulk temperature to a temperature where the defective B M T is dielectrically
lossy and would couple with the microwave field. This might account for the second
region o f maximum coupling. Since no other phases that could be responsible for this
second region are present, this explanation seems to be rational.
6.3.4 M icrowave synthesis o f B M T using BaO and Ta:0<.x
In an attempt to further investigate the Ta:0 5.x reaction. BaO was also used as a
reactant. Though water likely does not influence the synthesis above 290°C . the use o f
anhydrous BaO would eliminate the question o f T a i0 5.x oxidation. Table 6.3 gives the
X R D results o f the synthesis experiments using BaO. These results indicate that the
reaction progresses analogously to the B a(O H ): case in that both reactions result in the
formation o f similar products and both cease to couple once B M T is the major phase.
Hence, both synthesis routes proceed analogously with respect to microwave absorption
and consequent heating.
Additionally, since both reactions behaved analogously,
powder reactivity played no role in the results. It has been observed that a fairly reactive
system shows a tendency toward greater microwave absorption than more stable
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
132
systems, leading to faster heating rates. Fang27 studied the effect o f powder reactivity
and determined that there is a correlation between powder reactivity and microwave
absorption: reactive powders gave better results in microwave processing than more
stable precursors.
The heating rate plots are not sensitive enough to allow determination o f powder
reactivity or stability. However, microwave dielectric measurements performed during
various stages o f the reaction would give the loss factor (tan 5). which could indicate if
powder reactivity was a factor.
Table 6.3. X R D peak intensities from the B aO -M gO -Ta;0<.Kreaction
Condition
X R D Phase Identification and Relative Peak Intensities
Temp/Soak
1 32^5
BaO
MgO
BMT
3 6 4 T /0 min
100
46
13
20
367uC /0 min
18
479uC /0 min
22
11
9
100
497“C /0 min
26
18
14
100
13
'
"100
6.4 Conclusion
B M T could not be synthesized using a B a C 0 3 route. The mixture heated up to
1100°C basically showed no reaction, even using o f SiC as a secondary coupler in the
microwave system.
Experiments attempted with Ba(O H ): and BaO as reactants with
Ta; 0 5.K showed synthesis o f B M T .
Even though no experiment without SiC as a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
secondary coupler reached 800°C , it was determined that B M T was the major phase in a
partially reacted system.
Experiments involving SiC as a secondary coupler were more successful in B M T
synthesis.
Experiments attempted where the bulk temperature was 1200°C showed
almost complete reaction.
from a cold start.
Additionally the reactions were completed in a 20 minutes
Hence this method shows possible microwave enhancements over
conventional techniques.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
134
Chapter 7
M IC R O W A V E S Y N T H E S IS OF S IA L O N
7.1 Introduction
7.1.1 General
The term “S iA lO N ” was adopted in the early 1970s for a new group o f solid
solution compositions which contain Si, A l, O and N as major constituents and which
show promise as engineering materials for high temperature applications.
Silicon
aluminum-oxynitride (S iA lO N ) was first reported by Jack et al."w and Oyama et al. in
1971110.
It comprises a series o f compositions that are the result o f oxygen and
aluminum substituting on nitrogen and silicon sites in the Si3N 4 structure111112. Either
low- or high-temperature forms o f Si3N 4, a -S i3N 4 and P-Si3N 4, respectively, can form
S iA lO N materials by aluminum and oxygen substitutions111112. P-SiA lO N is the phase
most studied, and it can be prepared from several routes and phases111112. The phase
diagram for this system (Figure 7.1) shows that the material can be synthesized from
various mole percentages o f the end members because o f the wide stability field o f the
P -S iA lO N solid solutions.
The general formula for the P-phase is Si().iA l2O zN 8. / i:.
where z is the molar content o f substituted A l and 0 ions. The maximum value o f z that
would still maintain the P-SiA iO N structure is 4 .2 112. The general formula for the a phase is M pS i|: .(m»n)A l(m+n)O nN 16.n , where M =m etai ion (usually Y~3) 113. Hence, the a phase is stabilized by addition o f another cation112-113.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
135
Al
AUO:
AIN
SiO;
N
Si
Equivalent % N — *-
Figure 7.1. Phase Diagram for the S i-A l-O -N System
(reference111)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
136
In addition to the a - and P-phases, SiA lO N s have several other compositions. 0 ’phase S iA lO N s114 are the Al-substituted derivatives o f silicon oxynitride (S i:N :0 ) with
the general formula o f SU.jA I jO i ^ N i .*. O'-phase S iA lO N is stable near Si; N :0 in the
phase diagram. Another S iA lO N
phase that has been reported
is the X-phase
(3 A U 0 3 2Si3N 4) IIU15, which is silica-rich compared with other S iA lO N phases.
It
normally occurs as a minor constituent in P-phase production" ’. Several other SiAlO N
phases have also been reported by various workers111,112,116117.
7.1.2 Applications
P -S iA IO N is interesting because o f its strength and oxidation-resistance111,112.
SiAlO N s show promise as metal cutting and forming tools, as well as refractories.
Additionally. SiAlONs are being used as composite materials with S iC 118,1 |l> and are
under investigation for use as thin membrane filters120.
SiA lO N s have also been in use for metal wire- or tube-drawing tools for nonferrous alloys, including aluminum and copper. SiAlONs that maintain contact with
these metals at high temperature for long periods o f time show no reaction in the
absence o f fluxing agents. No interfacial reaction is observed between the metal and the
S iA lO N phase.
Dies made o f S iA lO N for copper alloys have shown good wear-
resistance compound with the dies currently used. However, i f the metal has oxidized
somewhat to form Cu20 . a reaction is possible with the SiA lO N material.
Hence,
avoidance o f oxidation is an important parameter.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
137
M etal cutting tools arc used for several attrition and wear applications in the
forming o f metal components112. S iA lO N materials have excelled in these areas because
o f toughness and thermal shock resistance.
tasks in the 1950s.
A130 3 tools were used to perform these
However, with the low toughness that is associated w ith A l:0 ,
materials, composites and Si3N 4 materials have become quite useful.
S iA lO N s also exhibit good creep- and environmental-resistance.
However.
S iA lO N , like many other ceramics, exhibit low fracture-toughness that may prohibit use
in demanding engineering applications118.
SiC whiskers have been added to A U 0 3
composites to improve fracture toughness.
Since SiAlONs have better mechanical
properties than A130 3, P-SiA lO N may be useful in such an application.
7.1.3 Synthesis
P -S iA lO N synthesis from a solid state reaction occurs at temperatures in excess o f
1 70 0°C li: for several hours.
compounds
Though S iA lO N has been synthesized from a variety o f
using solid state reaction, methods such as carbothermal
reduction
i is.i I9.i20.i2i.i22, ancj So|-gel123 methods have also been investigated.
Carbothermal reduction techniques have been employed by several researchers
i is.i I9.i20.i2[,i:2 jn ^
carbon.
conventional synthesis o f SiAlONs by nitriding various clays with
Lee and Cutler121 were the first to attempt nitridation o f a kaolin-carbon
mixture which produced SiAlO N s and other phases.
This is an economical way o f
producing S iA lO N because clay and carbon are inexpensive starting materials121.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
138
Since clays, carbon and some o f the intermediate products, such as silicon carbide
l23. are capable o f coupling w ith a microwave field very effectively, microwave
synthesis o f S iA lO N is very attractive. One possible reaction mechanism for P-SiAlO N
synthesis is given in equations 1-3. In a reaction attempted by Higgins and Henry123 to
form P -S iA lO N from kaolin and carbon, silicon carbide is formed as an intermediate
phase and serves with carbon as a reductant in the reaction.
It is proposed that first
kaolin dissociates into mullite, silica and water as shown in equation 1 and then the
silica is reduced by carbon to give SiC and CO (Eq 2).
Finally, mullite isreduced by
SiC and C in the presence o f nitrogen to give P -S iA lO N and CO (Eq 3).This
type o f
reaction should be valid for many clay materials, since mullite and silica are the first
phases formed.
3 (2 S i0 :- A I:0 ,- 2 H 20 ) = ( 3 A U 0 3- 2 S i0 : ) + 4(S iO ; ) + 6 (H :0 )
(1)
4 (S i0 2) + 1 2 ( C ) = 4{SiC) + 8(C O )
(2)
(3 A U O j-2 S iO :) + 4(SiC) + 3(C ) + 5 (N :)= 2(Si3A I30 3N 5) + 7(C O )
(3)
The systems studied for P -S iA lO N synthesis via carbothermal reduction were claycarbon-nitrogen. and boehmite-silica-silicon carbide-carbon-nitrogen. Each system was
processed by microwave and conventionally under nitrogen and the results were
compared.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
139
7.2 Experimental
7.2.1 Clay-carbon-nitrogen system
The clay chosen in this work was a montmorillonite (Aldrich Company. 5 pm
particle size) powder. This powder contained mica, kaolin and quartz as determined by
XRD.
However, the silica-to-alumina ratio matched that o f montmorillonite.
The
powder was mixed with graphite, giving a carbon-to-silica ratio o f 3 i:4. The mixture
was w et-m illed overnight and dried initially in a microwave oven and in a conventional
furnace for 16 hours at 120°C. The powder was uniaxially pressed into pellets, using a
Carver Lab press, with weights o f approximately 12 grams, diameters o f 2.873cm and
heights o f 1.016cm for all conventional and microwave experiments. A ll pellets were
fired in nitrogen atmosphere (Linde. Inc.).
7.2.2 A IO O H -S iO r S iC -C -N : (BSSCN) system
Boehmite and amorphous silica powders (Condea, Inc.. 0.5 pm particle size) were
mixed w ith silicon carbide and graphite powders. Samples were uniaxially pressed
pellets with weights o f approximately 12 grams, diameters o f 2.873cm and heights o f
1.270cm for all experiments. Sample treatment is similar to that given in the previous
section.
7.2.3 S ijN ^ A L C ^ system
Silicon nitride (H . C. Starck, Inc.) with a particle size o f <0.5 pm and alumina (lot
A16SG. Alcoa. Inc.) with a particle size o f 0.2 pm, were first mixed in equimolar
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
140
amounts, then were allowed to w et-m ill overnight. The pellet heights were about
0.762cm. Samples were heat-treated as in the previous systems.
7.3 Results and Discussion
7.3.1 Synthesis using Clay-Carbon-Nitrogen
Table 7.1 gives the X R D data for the microwave synthesis o f P -S iA lO N from clay
materials and graphite in a nitrogen atmosphere. The information in Table 7.1 shows the
relative intensities o f identified phases under the given time and temperature conditions.
Table 7.1. X R D Results o f the Microwave Clay-Carbon-Nitrogen Reaction
Conditions
Phases Identified by X R D and Relative Intensities
Temp/Soak
SiC
C
1300uC/2min
28
100
1400uC/15min
100
2
1400uC/30min
100
0
1500uC/15min
100
3
1500uC/30min
100
P -SiA lO N
3
—
r
j
J-S1A10N
mullite
other
10
Si'0:-I2
7
e S iA lO N -11
8
cSiA lO N -6
8
3
A IN -2 0
The data show that very small amounts o f SiAlO N phases were formed beginning at
1400°C.
The presence o f graphite indicates that the reaction has not gone to high
enough temperatures or for long enough time for completion.
However, it was not
possible to exceed 1500°C or remain at 1500°C for long periods o f time with the present
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
microwave equipment. The conventional synthesis results, shown in Table 7.2, indicate
no P -S iA lO N formation even after 48 minutes at 1600°C and only m inor amounts o f the
J -S iA lO N was identified by X R D .
Another major indication o f the incompleteness o f the reaction in microwave and
conventional synthesis was the large amounts o f SiC intermediate formed. This results
from the reaction o f carbon with SiO i. since a reduction o f the amount o f carbon is
observed along with an increase in the amount of SiC in both microwave and
conventional synthesis.
SiC acts as a reductant with carbon and finally reacts with
mullite forming SiAlO N (see equation 3).
Table 7.2. X R D Results o f the Conventional Clay-Carbon-Nitrogen Reaction
Conditions
Phases Identified by X R D and Relative Intensities
J-SiAlON"
mullite
other
10
S iO i-4
Temp/Soak
SiC
C
1500°C /30m in
15
100
1600uC /30m in
100
26
2
S i0 ;-4
16 0 0 uC/'48min
100
23
6
S iO :-6
P-SiAlO N
7.3.2 Synthesis using Boehmite-Silica-SiC-Carbon-Nitrogen(BSSCN)
Since SiC is an intermediate phase in the carbothermal reduction process the
addition o f SiC as a reactant might aid in the microwave synthesis o f S iA lO N . Further,
the presence o f SiC in the reactant mixture may increase the microwave absorptivity o f
the system.
Due to the complexity o f phases in the clay powder system chosen (see
with permission of the copyright owner. Further reproduction prohibited without permission.
142
experimental
section),
boehmite
(A lO O H )
and
amorphous
silica
mixture
was
investigated for the P-SiAlON synthesis.
Tables 7.3 and 7.4 report the X R D
results o f microwave and conventional
synthesis, respectively. The information in Tables 7.3 and 7.4 can be interpreted
analogously with Tables 7.1 and 7.2, with respect to the relative intensities o f the
identified phases.
Table 7.3. X R D Results o f the M icrow ave Reaction o f BSSCN System
Conditions
Phases Identified by X R D and Relative Intensities
c
Temp/Soak
SiC
1 100uC /30m in
100
1200bC /15m in
100
1
1300uC /15m in
100
13
1400uC /15m in
100
18
3
1500uC /15m in
100
30
4
P -S iA lO N
J -S iA lO N ’
mullite
other
33
81
2
Table 7.3 shows that more P -S iA IO N has been produced by microwave synthesis
from the BSSCN system than from the clay-graphite system mentioned in the previous
section.
The reaction nucleates at about 1200°C after a 15 minute soaking time.
In
conventional heating, the J-phase nucleates at 1500°C after 30 minutes soaking time and
the P-phase only at 1600°C (Table 7 .4 ). These results indicate that P -S iA lO N synthesis
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
143
has been accelerated in a microwave field and the minimum temperature for formation
was reduced to 1200°C from 1600°C in conventional heating.
Table 7.4. X R D Results o f the Conventional Reaction o f BSSCN
Conditions
Phases Identified by X R D and Relative Intensities
1 emp/Soak
SiC
c
1500“C /30m in
100
1600uC /30m in
1600uC /48m in
J-SiAION
mullite
29
10
94
100
25
15
100
rs
P-SiAlO N
3
16
other
"W """
88
Both microwave and conventional synthesis attempts have led to the formation o f
mullite first, then SiAlO N as temperature is increased. Since SiC was present in the
reaction mixture, and diffracts x-rays strongly, it is not possible to say if the
disappearance o f carbon is due to reduction o f mullite to the S iA lO N phase or the
reduction o f S i0 2 to produce more SiC. However, since SiA lO N phases are formed in
larger amounts than in the clay-carbon system, it may be reasonable to assume that
addition o f SiC as a reactant does help in the formation o f SiA lO N .
Currently it is difficult to directly compare the clay-carbon system to the BSSCN
system from the standpoint o f A120 3 and S i0 2 sources. Both syntheses were attempted
by different reactant routes.
However, the mullite phase is initially present in both
systems and S i0 2 phases are present in the clay-carbon reaction.
Further,
the initial
presence o f SiC in the clay-carbon system indicates the presence o f additional SiO: .
The presence o f mullite in both systems might indicate that the source o f A120 3 and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
144
S i0 2 is less important, since no S iA lO N formation occurs until A1:0 3 and SiO : phases
are first formed.
P -S iA lO N was not the major phase in any o f the reactions using either system.
This may be due to the complexity o f the carbothermal reduction process1" 125. There
are several factors that influence these reactions. Tw o o f the influencing factors are the
flow rate o f nitrogen and the size o f the pellet used1" 125.
Dijen and Metselaar125
showed that reaction rates could be enhanced by increasing the rate o f gas flow and
reducing the pellet size. Cho and Charles122 found that too high a flow rate leads to the
formation o f phases such as A IN .
affect the reaction rate.
Particle size and green density were also found to
Hence. P -S iA lO N synthesis using carbothermal reduction
methods involves nontrivial reactions that must be optimized by varying reaction
parameters and conditions.
7.3.3 Synthesis using silicon nitride and alumina
The diphasic reaction o f Si3N 4 and A120 3 to produce P-SiA IO N was tried first in
this investigation and remains the most successful.
The reason for investigating the
clay-carbon and the BSSCN systems was that the reactant phases would readily couple
with the applied microwave field. Clays (water-containing), graphitic carbon (electronic
conduction) and SiC (semiconducting) have microwave energy absorption properties
that are attractive for microwave processing. The Si3N 4 and A l;0 3 reactants have lowdielectric constants and neither is electrically conducting, so should not couple
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
145
effectively with a microwave field.
But contrary to this hypothesis, this system
produces P -S iA lO N after 15 minutes at 1500°C.
Tables 7.5 and 7.6 below report the X R D results for this system. The information in
these tables can be interpreted analogously with the data presented in other tables in this
section, with respect to the relative intensities o f the identified phases.
Table 7.5. X R D Results o f the Microwave Reaction between Si3N 4 and A l;0 3
Conditions
Phases Identified by X R D and Relative Intensities
P-S i 3N 4
1300uC /15m in
65
7
1300uC /45m in
60
7
1400°C /15m in
50
100
5D
l4 0 0 uC /30m in
58
6
100
1400°C /45m in
64
7
100
1400uC /lh r
55
6
"'100
1500uC /15m in
2
>
a -S l3N 4
O
Tem p/Soak
P-SiA lO N
x -S iA iO N
100
"
100
1
...
...
100
12
Synthesis o f P-SiA lO N was nearly complete after 15 minutes at 1500°C (Table 7.5)
using microwave heating.
temperatures > 1500°C.
Conventional synthesis nucleates the S iA lO N phase at
However, comparing all the conventional synthesis methods,
more P -S iA lO N was produced from the Si3N 4-A l; 0 3 system than from any other
reaction route. This is likely due to the structural similarity between P -Si3N 4 and P-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
146
Table 7.6. X R D Results o f the Conventional Reaction between Si3N 4 and A l:0 3
Conditions
Phases Identified by X R D and Relative Intensities
Temp/Soak
P-S i 3N 4
a -Si3N 4
1500uC/15m in
54
6
100
1500uC/30m in
58
7
100
1600uC/18m in
57
1600uC/30m in
100
39
l
59
100
77
8
1648uC/48m in
54
100
84
12
1700UC/1 hour
22
75
100
25
c
T-SiAlON
>
P -S iA lO N
S iA lO N 111. P-SiAlON is formed when A l- 0 components replace Si-N in the P-Si3N 4
structure
to obey the general
formula
S i ^ A l ^ N * - / 111' 112-
thermodynamically favorable to make P -S iA lO N
Hence
it may be
from the S i3N 4-A l20 3 route as
opposed to the carbothermal reduction or other routes.
7.3.4
Microstructure analysis o f Si3N 4-A l20 3 route
The microstructure o f P-SiAlO N synthesized from the S i3N 4-A l20 3 route shows
differences for microwave and conventional synthesis.
Figures 7.2 and 7.3 are SEM
micrographs o f the products from the microwave reaction after 15 minutes at 1500°C
and from the conventional reaction at 1648°C for 48 minutes, respectively. Figure 7.4 is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
147
Figure 7.2. S E M Micrograph o f the Products from the Microwave Reaction between
Si3N 4 and A1:0 } at 1500°C for 15 minutes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
148
08
-------------1 0Mm
15KU
X2.O00
14mm
Figure 7.3. SE M Micrograph o f Products from the Conventional Reaction between S i,N 4
and A l:0 3 at 1648°C for 48 minutes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
149
SIA LO N
CBM
Figure 7.4. SE M Micrograph o f Products from the Conventional Reaction between S i,N 4
and A1:0 3 at 1700°C for 1 Hour
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
150
an S E M micrograph o f the conventional reaction heated 1 hour at 1700°C. The
submicron-sized grains o f the microwave-reacted sample appear uniform.
The
conventionally reacted sample appears multiphasic, which agrees with the X R D results.
The presence o f long column-like structures in the center o f the sample are evident.
These structures have sizes in the 30-40pm range.
E D A X analysis identified these
structures as P-SiAlON.
7.4 Summary
It was difficult to make any S iA lO N from the clay-carbon reaction or the BSSCN
system. From the X R D data it appears that higher temperatures and/or longer times are
needed to completely synthesize S iA lO N by microwave or conventional methods. This
is rationalized by the fact that SiC, an intermediate phase, still remains in large
quantities before P-SiA lO N is formed.
Reaction parameters must also be controlled
more precisely with respect to pellet size, nitrogen flow rates and reactant quality, to
make the P-SiA lO N phase.
However, using A120 3 and S i,N 4 precursors. P-SiAlON
was microwave synthesized in 15 minutes at 1500°C. Conventional P-SiAlON reaction
occurs at higher temperatures (>1500°C ) and longer times.
submicron-sized
uniform
grains
for
the
SEM analysis shows
microwave-processed
sample.
The
conventionally processed sample has a diphasic microstructure with one o f the phases
having considerably larger grain size. The mechanism for this synthesis likely involves
P -S iA lO N being formed by A l- 0 substitution for Si-N into the P-Si3N 4 structure.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
151
Further work is needed to synthesize P-SiAlO N with boehmite or clays and Si3N 4 to
provide initial microwave absorption for the system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 8
M IC R O W A V E S Y N T H E S IS OF PZT
8.1 Introduction
8.1.1 General
Lead zirconate titanate. Pb(Zrx T i | . , ) 0 3 or PZT, is one o f very few high volume
electroceramic materials having the perovskite structure o f A B O / * 89. Those aspects,
including ferroelectricity, were discussed in Chapter 5. Pb'2 ions occupy the A-site and
the B-sites are occupied by Zr*4 and T i*4. The extent to which Z r or Ti occupies the Bsite is dependent upon the formulation o f the starting reactants, since both ions have the
same charge and roughly similar sizes. The phase diagram (Figure 8 .1) between the end
members P b Z r0 3 (P Z) and P b T i0 3 (PT) shows complete solid solubility89. The cubic
phase, whose boundaries change with specific composition, is the paraelectric phase.
Hence the cubic phase boundary represents the Curie temperature (T c) as discussed in
Chapter 5. The morphotropic phase boundary, at the composition x= 0.48. separates the
tetragonal phase from the rhombohedral phase89. The PZT composition that has shown
the highest piezoelectric properties89 has an empirical formula o f Pb(Zr0 52 Ti,M8) 0 3.
The properties o f PT-PZ ceramics can be altered by doping with donor elements such as
La, Nb. Sb or W , or by acceptor elements such as Fe. Mn. N i and Co.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
153
500
Curia-tam paratura
Cubic
.1____
400
300
Tatragonal
“
200
Morphotropic
phase
boundary
J-------------
100
0
10
20
30
40
50
60
Pt> riO ,
70
80
90
100
PbZrO,
mol V P b Z rO j
»
Figure 8.1. The Binary Phase Diagram for the P b T i0 3-P b Z r0 3 System (reference*')
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
154
8.1.2 Applications
Since the 1950s. PZ-PT materials have been commercially exploited because o f
their excellent piezoelectric properties86'89 126. Directly after sintering, these materials
are not piezoelectric; they must be poled with an applied electric field86'89 greater than
the coercive field in order to induce polarization. This is accomplished by heating the
material in the 100-200°C range and then cooling it while applying the electric field
during the entire process.
These materials are used as transducers, in sonar and medical applications, by
receiving or transmitting elastic energy into liquids or solids. Piezoceramics have been
used in ultrasonics, which require higher frequencies affordable by these materials.
These materials convert electrical energy into mechanical energy or vice-versa89. PZT
has also been used in delay line transducers. Data systems, such as computers, have
need to delay signals by specific amounts o f time (increasing from lpsec)86.
Sound
propagation, in the form o f square wave pulses or bursts o f sinusoidal waves, through
liquid and solid media can provide adequate distances for time intervals. The media
used may be a metallic strip, fused silica or liquid mercury.
High frequency use
involves silica rods with piezoelectric materials bonded to them. High piezoelectric
coupling gives improved performance over silica materials; hence, P Z T has replaced
silica in this application for frequencies up to 20M H z.
For frequencies substantially
higher than 1M H z. the dielectric constant must be low and the transducers should be
2500 m/sec) is used.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
155
8.1.3 Synthesis Techniques
PZT has been successfully synthesized by several techniques including solid state
I27.i2*.i29 mjcrowave soiij sjate processing127 129 130, hydrothermal techniques131, and
sol-gel methods127 132 133. PZT is conventionally synthesized from PbO or Pb30 4, Z rO ;
and T iO : powders. P b C 0 3 has also been used as a Pb source in many Pb-based material
synthesis processes because it has a lower vapor pressure than PbO.
PbO loss during
the synthesis o f PZT is a common problem since partial vaporization o f PbO from the
reactant system w ill allow for the formation o f phases other than stoichiometric PZT 134
and adversely affect the dielectric properties. Using PbO as a Pb source. Ounaies et
a l.127 prepared phase-pure PZT at 800°C for 4 hours after grinding the reactants in Z rO :
media for 24 hours with ethanol.
Below 800°C'. PbO. ZrO : and PbTiO , phases were
present with PZT.
Microwave solid state synthesis o f PZT was reported by Ounaies et a l.127. They
microwave synthesized PZT using the same reactants used in the conventional
synthesis. However, in contrast to the conventional method, they found that phase-pure
PZT could be synthesized in a microwave field after 45 minutes at 720°C.
temperature increased rapidly after 200°C . indicating thermal runaway.
sizes observed in samples made by both methods were similar:
The
The particle
however, the
microwave-processed material had a more uniform particle size distribution.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
156
In a microwave-hydrothermal system, PZT and many other perovskite phases were
synthesized by Komameni et al., using aqueous solutions o f P b fN O }): , Z rO C I: and
T iC l4 in K O H . They used a microwave-hydrothermal technique using apparatus that
operated at 2.45GHz. The maximum pressure achievable was 200 psi.
Due to the use
o f microwave energy. Teflon Parr-bombs were used instead o f metallic Parr-bombs.
It was found that PZT was the only phase present after the microwave experiments
carried out in the temperature range o f 115°C-164°C for 0.5-1 hour. A 2-hour soaking
period at 164°C resulted in P Z T and PbO formation. It was also found that P Z T could
not be synthesized by this route by a conventional-hydrothermal method until after 24
hours in the temperature range o f 138°C-164°C.
Sol-gel processing o f PZT has been reported by several researchers12713:133.
Microwave and conventional sol-gel techniques were compared by Ounaies et al.. using
lead
acetate
trihydrate,
zirconium
isoproproxide and strontium metal.
n-isoproproxide
(70%
in
butanol),
titanium
2-methoxyethanol was used as the solvent.
First
lead acetate trihydrate was dissolved in the solvent and dehydrated at 120°C.
Sr-metal
was
titanium
added
after
cooling
to
70°C.
Zirconium
n-isoproproxide
and
isoproproxide were added and the resulting solution was refluxed at 70°C for 12 hours.
This solution was hydrolyzed with water and ethanol and allowed to gel in a ventilated
oven at 70°C. The gel was calcined conventionally at 700°C and 6 00 cC for 4 hours and
5 hours. Microwave calcination was done at 600°C for 40 minutes. In each case phase
pure P Z T was obtained.
However, P Z T was synthesized by microwaves at lower
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
temperatures and times. M icrowave synthesized PZT powders had an average grain size
o f 2pm . The conventional synthesis showed an average grain size o f 4pm.
8.2 Experimental Procedure
8.2.1 Microwave synthesis o f P Z T using reduced TiO :
P Z T reactant mixture was prepared by mixing chemically stoichiometric amounts
o f P b C 0 3 (Hammond), ZrCK (M allincrodt), and T iO :.v T iO ; .x was reduced by the
method described in Chapter 4. The mixture was then dry m illed, to prevent oxidation
o f T iO :.v using ZrO ; media for 20 hours in a small polypropylene container. The ballm illing speed was approximately 55 rpm. The media were separated from the reactant
mixture by passing it through a 40-mesh screen.
In order to increase particle contact, powders were pressed into pellets using a
stainless steel die. The mixture was pressed uniaxially into pellets using a Carver Press
at 10,000 lb. The die had a diameter o f 1.270cm. Each pellet weighed approximately 6
grams and had heights o f approximately 0.762cm.
The 5 -K W multimode microwave unit was used in this work.
placed in a covered alumina crucible on A l20 3 fiber insulation.
The pellets were
The thermocouple
directly contacted the sintered alumina crucible from underneath. N 2 gas flowed from
the bottom o f the crucible through the alumina tube and exited from the top o f the cavity
through air vents. Prior to each experiment the system was purged for 30 minutes.
The power used to microwave process the samples was
experiments.
No secondary coupling technique was used.
1.2 K.W for all
The heating rate o f the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
158
samples was not controlled. Each sample was analyzed by X R D for phase information
and by S E M for microstructural information. No E D A X data was obtained.
8.3 Results and discussion
8.3.1 Microwave synthesis o f PZT
Table 8.1 reports the X R D data obtained in the microwave synthesis o f defect-PZT
from T i 0 2.x.
PZT formation was near completion by 600°C, with only minor amounts
o f the reactants present at this temperature.
Experiments carried out at higher
temperatures indicated that by 900°C only trace amounts o f Z rO : and PbO were present.
W ith the dry-milling technique used, the reactant mixture may not have been well
mixed, which could explain the presence o f minor amounts o f reactant phases after heat
treatment.
Table 8.1. X R D peak intensities from the P b C 0 3-Z rO ; -T iO : synthesis
X R D Phase Identification and Peak Intensities
T iO :
600uC /0m in
4
700uC /0m in
1
800JC /0m in
1
900uC /0m in
PbO
cub-PZT
tet-PZT
2
5
100
45
3
2
5
100
100
2
2
4
100
2
100
ZrO,
1
o
u*
Temp/Soak
CT
Condition
”
100
100
However, the use o f reduced T iO : in the microwave synthesis o f PZT has
drastically lowered the temperature and/or time conditions necessary to synthesize PZT
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
159
as reported by previous workers127129. The total time necessary to achieve 600°C was 6
minutes using 1.2 K W o f output power. X R D results indicate that the cubic PZT phase
formed first, followed by the tetragonal phase. This appears to be the opposite o f what
occurs under equilibrium conditions. According to the phase diagram (Figure 8.1) the
tetragonal phase is stable (for the composition used in this work) up to about 360°C.
above which the cubic phase is stable. A t present, the existence o f a cubic phase o f nonstoichiometric P Z T at low temperatures is not known. It is believed that this may be a
metastable condition due to the rapid heating in a microwave field.
In the case o f microwave-synthesized defect B a T i0 3 (Chapter 5) the hightemperature hexagonal phase was formed first, followed by the tetragonal phase which
is the thermodynamically stable phase in the temperature ranges reported. However, the
hexagonal phase can be stabilized by structural or lattice defects.
Hence, it may be
possible for defect structure stabilization o f the cubic PZT phase at low temperatures.
8.3.2 Microstructural analysis o f PZT
S E M micrographs (Figures 8.2, 8.3 and 8.4) o f the PZT sample heated to 800°C
show the initial stages o f sintering at various points o f the compacted pellet. In Figure
8 .2 . triple points as well as conglomerations can be observed in the microstructure.
Figure 8.3
shows
further evidence o f grain growth and grain conglomeration.
Additionally, the presence o f a liquid phase is also apparent from the curvature in the
impressions left on the faces o f certain grains by other grains as they pull away. Figure
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
160
8.4 shows curved impressions on several o f the grains and possible “ necking” at the rim
o f these impressions, which is an indication o f a liquid phase.
8.4 Summary
P Z T has been synthesized using P b C 0 3, Z r 0 2 and T iO :.x powders in a microwave
field at temperatures as low as 600°C . The cubic phase was found to nucleate first,
followed by the tetragonal phase. M inor amounts o f the reactants remained even after
heating to 900°C.
This may be due to the mixing heterogeneity that is intrinsic with
dry-m illing as opposed to wet-milling and sol-gel techniques. The microstructure o f the
sample heated to 800°C shows evidence o f sintering and the possibility o f the presence
and disappearance o f a liquid phase.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
161
Figure 8.2. SE M Micrographs o f P b C 0 3, ZrO , and T iO :„ Microwave Heated to 800°C in
Nitrogen Showing the Beginnings o f Sintering
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
162
H I-------_
4 8 4
1 5KU
i0 M m
X 2 / 5 0 0
15mm
Figure 8.3. SEM Micrographs o f PbCO„ Z rO ; and T iO ,.t Microwave Heated to 800°C in
Nitrogen Showing Grain Growth
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
163
Figure 8.4. SE M Micrographs o f P b C 0 3, Z rO : and T iO ,., M icrowave Heated to 800°C in
Nitrogen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
164
Chapter 9
O TH E R S Y S T E M S IN V E S T IG A T E D
9.1 Introduction
In addition to the systems reported in the previous chapters. limited attempts were
made to microwave synthesize two other systems. Zircon synthesis was attempted using
Z rO : and S i0 2. T iB : synthesis was attempted using TiO : .x and B 4C. T iO ; .x was found
to absorb microwave energy very efficiently and aid in accelerating the synthesis of
several materials as well as change the reaction mechanism for certain systems as
reported in the previous chapters.
systems.
However, all o f that work was done on oxide
Hence, it would be interesting to observe i f T iO ;.x would show similar
enhancements in reactions using a non-oxide component.
9.2 Zircon (Z r S i0 4)
Zircon is a naturally occurring mineral having a Mohs scale hardness o f 7.5. Its
primary occurrence is in SiCK-rich igneous rocks such as granite135.
However, it is
rarely found in rocks in economically mineable quantities135. Economic concentrations
o f the mineral are found after granitic rocks have been weathered and eroded135.
Zircon can be formed experimentally from a solid state reaction between Z rO ; and SiO:
in the temperature range of 1400-1600°C136137. At temperatures above 1600°C. zircon
dissociates into its constituent oxidesl37.
Zircon is used for several industrial purposes.
It has been used as a refractory
material138 139 due to its high melting or dissociation temperature, low
thermal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
165
expansion coefficient 140141 and high thermal shock resistance. Zircon is also used in
porcelain materials, pigments and T V
phosphors141, as a Z rO ; source142 and in
com positesl43.
9.2.1 Experimental Procedure
9.2.1.1 Solid state synthesis
In this work, monoclinic Z rO : (T A M , Inc) having a particle size o f 5pm was used.
The S i0 2 source was a quartz powder (U.S. Silica. Inc.) having a particle size
distribution o f 5- 10pm.
Equimolar amounts o f the precursors were wet-milled
overnight using ZrO: milling media. The resulting mixture was heated in a microwave
oven at low power levels for 1 hour to near dryness, then in a conventional oven
overnight at 110 °C .
9.2.1.2 Sol-gel synthesis
Z rO : source was zirconium n-proproxide (ZP). which is 70% in propanol (Alfa
Products). The SiCK source was tetraethylorthosilane (TE O S) (Aldrich Chemical Co.).
The procedure o f Barrie and Meshishnek [ l44] was used for sol preparation such that the
chemical stoichiometry o f Z rS i0 4 is maintained. A 1:1:8 molar ratio o f TEOS. ZP and
ethanol was stirred for one hour and added dropwise to a 6:6:24 molar ratio o f H 20 .
12N N H 4O H and ethanol. The entire mixture was stirred and allowed to gel. The gel
was dried by microwave oven and later calcined in a conventional fumace for 2 hours at
500°C.
In order to increase the particle contact area, the powders produced from sol-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
166
gel and solid state techniques were pressed into pellets using a Carver Lab press at a
pressure o f 9000-10,000psi. The powders were pressed in a stainless steel die having a
diameter o f 2.873cm.
The weights o f the pellets were between 7 -1 1 grams and the
pellet heights were about 0.762cm (solid state derived powders) and about 2.032cm
(sol-gel derived powders).
9.2.2 Results and Discussion
9.2.2.1 High temperature X R D analysis o f the ZrCL-SiCT reaction
The data presented in Table 9.1
were obtained in high temperature X R D
experiments. The conventional methods and temperatures required for zircon synthesis
were mentioned above.
However high temperature X R D was attempted to determine
exactly where crystallographic transformations and synthesis occur.
Table 9.1. High Temperature X R D Analysis o f Zircon synthesis
Conditions
Temperature
Phases Identified by X R D and Relative Intensities
Time
Zircon
......
ZRJ;
Z rO ,
(monoclinic)
(tetragonal)
Ambient
Initial
100
1300“C
2 hours
1
1350UC
1 hour
1
100
1400UC
2 hours
6
100
1450UC
2 hours
13
100
1450UC
4 hours
15
100
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
167
From the high temperature X R D data it can be observed that the conventional solid
state reaction does not go to completion even after 4 hours at 1450°C and in
experiments by other workers 137 28 hours or longer were required.
9.2.2.2 Microwave synthesis o f Z rS i0 4
Processing o f zircon in a microwave field proved to be a very difficult task from a
solid state reaction or sol-gel route.
Secondary couplers were used because ZrO : or
S iO i do not effectively absorb microwave energy. However. Z r 0 2 or SiO: may begin to
absorb slightly when the temperature is high enough to permit dielectric loss.
Additionally, since ZrO ; has very poor thermal transport properties, it could quickly
undergo thermal runaway conditions.
Using SiC as a secondary' coupler, zircon could not be microwave synthesized.
Using Z rO ; board as a secondary coupler, some zircon was produced but temperature
could not be reliably measured.
“ Hot spots” developed in the Z rO : board, and
prevented the desired volumetric microwave absorption. Temperature was measured at
the center o f the sample surface, but the sides o f the sample (closest to the Z r 0 2 board)
were likely at temperatures considerably higher than center.
Table 9.2 shows the results o f the zircon synthesis using Z rO ; board. The data is
questionable because o f questionable temperature measurement.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
168
Table 9.2.
X R D Peak Intensities o f Zircon Synthesized by Microwave Furnace Using
Z rO i Board (N D -n o reliable data available)
1350°C
Tim e
1400°C
1450°C
1500°C
MW
Mw
Mw
Mw
15
min
2
8
18
22
30
min
6
2
2
ND
45
min
7
ND
ND
ND
1 hr
"" rciT
2
..
ND
94
9.3 Titanium diboride (T iB : )
T iB ; is a refractory material with a high melting point (near 3000°C) with a Knoop
hardness o f 3400 kg^mm: using a 500g load. It has been synthesized from methods such
as borothermic reduction o f T iO ; l45, thermite reactions 14b. mechanical alloying o f Ti
and B U7U8, and by reacting hydrolyzed titanium butoxide with boron powders l4<’.
Because T iB ; is oxidation resistant up to 1000°C and has low electrical resistivity l5°. it
has been investigated for use as coatings for cutting tools l51, wear resistant materials
l5°. ballistic armor l5:, and as a component in composites with SiC. T iC and A l: 0-,
l50153. T iB 2 has been shown to increase the mechanical strength and properties o f the
matrix phase.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
169
9.3.1 Experimental procedure
Chemically stoichiometric amounts o f B4C (Cerac, Inc.) and reduced T iO : were
ground and dry milled in a Nalgene™ container.
Dry milling was employed to prevent
the oxidation o f the T iO i., phase. After m illing, the powders were separated from the
media by vigorous sieving using a 40-mesh screen.
In order to increase the particle contact area, the powders were pressed into pellets
using a Carver Lab press at a pressure o f 9000-10,000psi. The powders were pressed in
a stainless steel die having a diameter o f 2.873cm using propanol as a binder.
The
pellets weighed about 11-12 grams and were about 1.016cm thick.
9.3.2 Results and discussion
The X R D results presented in Table 9.3 indicate that T iB ; can be synthesized from
this route. However, several additional phases are also present. This indicates that there
are several competing reactions during T iB 2 synthesis from T iO ;.x and B 4C.
Table 9.3 X R D peak intensities from the synthesis o f T iB : from T i 0 2.x and B4C
Conditions
Phases Identified by X R D and Relative Intensities
T A
Temperature
T iO :.x
B 4C
1000UC
29
29
22
48
26
l'DO
1 0 5 0 °r
6
42
19
12
....
TOO"
....
1100°C
10
100
60
1200UC
15
T IB ,
" '8 6
2
83
" T i:0 3
' " T iB O j
'
84
74
14
100
..
—
T
..
53“ "
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
170
N o obvious trend exists in the reaction as temperature increases, with respect to the
disappearance o f the competing phases. As T iB ; increases the other product phases also
seem to increase.
The existence o f T i20 3 and T i30 5 phases indicate that part o f the
T iO i.x concentration is reducing to form these phases, while other amounts react to form
T iB i and T iB 0 3. B4C reacts to form T iB ; , T iB 0 3 and H 3B 0 3. The existence o f H 3B 0 3
likely indicates an impurity. Since an organic binder was used with each pellet and the
process atmosphere was nitrogen gas, some o f the binder may not have evaporated or
oxidized, but reacted with B4C. The organic material was the only known source o f
hydrogen in the system.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
171
Chapter 10
G E N E R A L C O N C LU S IO N S
In this work powders o f several materials (A l2T i 0 5, B aTiO j, B M T . PZT. SiAlO N .
and zircon) have been synthesized using various precursor reactants in a microwave
field. Most synthetic attempts have been made via solid state reactions, which are the
most economically favorable and common reaction processes for industrial purposes.
M any o f the attempted syntheses o f powder materials in a microwave field were not
reported prior to this work.
Furthermore, a new synthesis technique using reduced oxide precursors has been
developed. The use o f non-stoichiometric or reduced phases such as T i 0 2.x and Ta20 5.x
in microwave synthesis has not been reported prior to this work. Using these reactants
in microwave processing, the synthesis o f A I2T i 0 5. B a T i0 3. B M T and PZT have shown
major enhancement o f the reaction kinetics over conventional routes as well as
stoichiometric microwave synthesis.
Product phases were found to form at much
shorter time and lower temperature conditions than would have formed under normal
solid state reaction techniques.
There maybe several mechanisms that explain the extraordinary ability o f nonstoichiometric phases such as T i 0 2.x and Ta20 5.x to absorb microwave radiation
efficiently and heat.
The two most likely mechanisms for microwave absorption in
these systems are electronic and ionic conduction. Since these metal oxides likely
contain oxygen vacancies, titanium aliovalencv is necessary to compensate for the
charge imbalance.
Hence in T i 0 2.x the existence o f T f 4 and T i* 3 ions is likely.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The
172
existence o f aliovalency may provide a mechanism for microwave absorption by
electrons hopping between T i*4 and T i* 3 ions or electronic conduction in response to the
application o f the electromagnetic field, which is the mechanism for absorption in
semiconductors.
Electron hopping or conductivity results in friction buildup in the
material which is responsible for heating.
Another microwave absorption mechanism maybe ionic conduction. Ti-ions, under
the influence o f an electrical field, may have the ability to move through the lattice.
This would cause a fiction buildup inside the lattice which would result in localized
heating accompanied by thermal expansion.
Expansion in the lattice would increase
diffusion o f T i through the lattice, possibly increasing electromagnetic absorption and
initiating thermal runaway.
In the case o f A l:T i 0 5< it was found that the microwave reaction nucleates the
phases faster but growth is slowed as temperature rises.
This is possibly due to
difficulties encountered by the T iO :.x as it attempts to diffuse across the growing
product interface layer. It may also be possible for T iO :.x to re-oxidize as temperature
increases.
This could result in decreased microwave absorption and lowered reaction
rates.
I f T iO :.x is reduced upon heating, and becomes more metallic or a stoichiometric
reduced phase such as T i:0 3 or TiO . it may cease to absorb microwave energy
effectively.
However, dielectric measurements (Chapter 4) made in a reducing
atmosphere (forming gas) suggest that i f T iO : .x reduces further it couples more
effectively. This probably indicates that T i 0 2.x heated in nitrogen has a smaller degree
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
173
o f defects or oxygen vacancies than T i0 2.x heated in forming gas. but is not reduced to a
metallic or stoichiometric phase. Hence, T iO :.x can be reduced significantly, according
to the dielectric data, and maintain the rutile structure.
In the case o f B a T i0 3, it was observed that the microwave synthesis using nonstoichiometric methods produces the phase by 700°C.
fast diffusion o f T i 0 2.x.
This is likely accomplished by-
This is a different reaction mechanism from the reaction
pathways for conventional diffusion.
The conventional pathway, as observed by
Beauger, et a l.96, shows BaO as the diffusing species.
The first phase to nucleate, at 250°C, was the high-temperature hexagonal B a T i0 3.
This
is
likely
because
the
hexagonal
phase
is
kinetically
favored
thermodynamically favored tetragonal phase by the presence o f defects.
over
the
However, as
temperatures reached 600°C the tetragonal phase nucleates and the hexagonal phase
disappears.
Additionally, it was determined that B a T i0 3 could be melted at 1000°C using nonstoichiometric methods with solid state precursors.
The microstructures in the SEM
micrographs conclusively show that B a T i0 3 phase had melted.
B M T (Ba(M g033Tafl67)O 3) was synthesized by a non-stoichiometric microwave
route. Only B a(O H )2 and BaO. used as barium sources, yielded B M T . No B M T phase
was synthesized using B a C 0 3.
SiC was needed as a secondary coupler for these
experiments, because even non-stoichiometric B M T does not absorb microwave energy
effectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
174
PZT
was also successfully synthesized by a non-stoichiometric
route
in a
microwave field. The reaction mechanism appeared to be similar to that o f B a T i0 3< in
that a kinetically favored phase nucleates before the thermodynamically favored phase.
Cubic P Z T nucleates at about 300°C, and as temperature increases, tetragonal PZT
nucleates.
A t 800°C both phases are nearly equal in proportion.
Microstructural
examination shows that sintering initiates around 800°C.
P -S iA lO N was successfully synthesized from Si3N 4 and A1:0 3 precursors.
microwave
fumace. P-SiAlON
was synthesized at
1500°C
in
In the
15 minutes.
In
conventional synthesis methods, the reaction was not complete by 1750°C after 1 hour.
Several attempts to synthesize P-SiAlON from a carbothermal reduction route produced
various S iA lO N phases at lower temperatures than the Si3N 4 and A l:0 3 route: however,
single phase S iA lO N was not obtained using this method.
It is rationalized that the
reaction conditions and parameters must be refined to successfully synthesize single
phase P -S iA IO N in a microwave field.
M icrow ave experiments to synthesize zircon were not successful. ZrCK and SiO;
do not couple with a microwave field at 2.45 G H z, making zircon synthesis very
difficult.
Attempts were made to process zircon using various secondary couplers to
compensate for the lack o f microwave absorption by the system. The systems used as
secondary couplers were zirconia fiber board, zirconia fiber board pieces and SiC rods.
The zirconia board insulation developed hot spots that provided conventional heat.
Hot spot generation causes temperatures to escalate nearly exponentially. The fact that
the processed samples were exposed to temperatures considerably higher than the set
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
point temperature was the reason for glass phases detected in the product. SiO : phase
apparently superheated and melted due to the high temperature generated by the
secondary coupler. SiC rods also provided conventional heating without hot spot
formation: however, heating rate plots show that the sample's heating rate approximates
that o f SiC, indicating that the sample did not absorb the microwave energy appreciably.
10.1 Suggestions for Future Work
The use o f “non-stoichiometric" microwave synthesis in a nitrogen, or inert,
atmosphere has proven to be successful in powder synthesis. However, from the
dielectric measurements carried out on T iO :.x in forming gas (chapter 4. Figures 4.1213). there is a possibility o f further enhancements in synthesis rates.
Attempts should
be made to synthesize these and other materials in a reducing atmosphere.
From the experiments carried out in this work it has been rationalized that the
diffusing and reacting species is the reduced metal oxide phase. It has also been shown
that different reaction pathways have resulted.
Hence, diffusion studies are needed to
further understand the kinetic pathways o f these systems.
This would particularly be
useful for B a C 0 3-T iO ; in air. It is currently rationalized that T iO ;.x is coupling with the
field and diffusing.
In the conventional case, BaO was found to be the diffusing
species96. Placing pellets o f B a C 0 3 and T i 0 2.x in contact and microwave reacting them
in nitrogen, and repeating that experiment using conventional heating may provide more
insight into the reaction mechanism.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
176
Additionally, dielectric analysis should be attempted on all non-stoichiometric
microwave reactions. Though it is apparent that the system is dielectrically lossy by the
existence o f thermal runaway, loss measurements would aid in determining the
conditions necessary to optimize the reaction process.
The true measure o f the success o f this novel technique might be in the ability to
sinter synthesized compounds. The microstructures o f B M T and PZT do show evidence
o f presintering.
that
w ill
However, the powders o f both materials were made using precursors
evolve
gases
during
reaction
and
prevent
sintering.
If
the
non-
stoichiometrically synthesized powders were compacted and microwave processed in
their reduced state, enhancements in sintering may be realized.
Additionally, investigation o f frequencies other than 2.45 G H z or processing using
several frequencies in the microwave electromagnetic region simultaneously, such as a
variable frequency microwave furnace154155, may also provide interesting results.
Benefits have been realized using frequencies other than 2.45 G H z for processing white
ceramics29.
Hence it maybe possible to provide enhancements in non-stoichiometric
microwave processing over what has been reported in this thesis.
10.2 Concluding Statement
Non-stoichiometric or defect microwave synthesis has proven to be a viable tool in
making powders o f commercial ceramic materials, in certain cases more so than
stoichiometric microwave processing.
The overall simplicity o f solid state synthesis
from metal oxides as opposed to synthesis from a sol-gel route, in addition to the
decreased processing times and temperatures, serve to make this an attractive processing
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
177
method.
The step o f making “defective” precursors is where time is consumed.
However, these materials, particularly T iO : .v are stable in this reduced state. Hence it
is possible to store these materials.
This would allow for large or industrial scale
synthesis.
Additionally, very little microwave power is needed to initiate or maintain the
reactions since these reduced precursors strongly absorb the applied energy. Hence this
process should show various savings in time and energy.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
178
R E FE R E N C E S
1Binner. J.G.P.. “Microwave Processing o f Ceramics”. Br. Cer. Proc., 45.97, (1989)
2 Sutton, W .D ., “Microwave Processing o f Ceramics-an Overview". Mat. Res. Soc.
Symp. Proc.,
269,3, (1992)
' Neas, E.D. and Collins, M.J.. “Microwave Heating: Theoretical Concepts and
Equipment Design”, Introduction to Microwave Sample Preparation: Theory and
Practice, H .M . Kingston and L.B. Jassie, eds.. American Chemical Society, Washington
D C , (1988)
»Roy. R., “ Fifty Years o f Mineral and Materials Synthesis”. /„ ,/ Ceram
Wonagr 1
[2]. 737. (1994)
5 Roy, R., Komameni. S. and Yang. L.J.. “Controlled Microwave Heating and Melting
o f Gels”..;
4
m Cer. S o c- 68 P I- 392- 0 9 8 5 )
* Sutton, W .D.."M icrowave Processing o f Ceramic Materials". C er B u ll • ^8 P I- 37().
(1989)
' Campisi. I.E ., Summers, L.K .. Finger, K.E.. and Johnson. A .M .. “Microwave
Absorption by Lossy Ceramic Materials", Mat. Res. Soc. Symp. Proc.. 269.157. (1992)
«Quemeneur. L. et al.,"Microwave Clinkering with a Grooved Resonant Applicator", j
Am. Cer. S o c- ^6 0 2 1> 833>(1983)
9 Metaxas, A .C . and Binner. J.G.P., "Microwave Processing o f Ceramics". Adv. Cer.
Proc. Tech., 1, 285, J.G.P. Binner, ed. Noyes Publications. N.J., U .S.A ., (1990)
|° Metaxas. A .C ., "Microwave Heating", p w r £ ng j „ 5 [5], 237, (1991)
" Newnham. R.E.. Jang. S.J., M ing, X . and F. Jones. "Fundamental Interaction
Mechanisms Between Microwaves and Matter". Cer. Trans.. 21, 51. (1991)
i: Wames, L .A .A ., Electronic Materials, 215-6. Van Nostrand Reinhold. N ew York.
N Y . (1990)
i’ Katz, J.D., Blake, R.D. and Kenkre, V .M ., "Microwave Enhanced Diffusion?”. Cer.
Trans., 21, 95, (1991)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
179
i* Kenkre, V .M .. Skala. L. and Weiser, M .W ., “Theory o f Microwave Interactions in
Ceramic Materials: The Phenomenon o f Thermal Runaway”, , / \ ( a l
26. 2483.
(1991)
■» Kenkre. V .M ., “Theory o f M icrow ave Interactions with Ceramic Materials". Cer.
Trans.. 2 1 .6 9 . (1991)
von Hippel, A., Dielectrics and Waves, Technology Press o f M IT and W iley. New
York, N Y (1954)
i’ von Hippel, A.. Dielectric Materials and Applications. Technology Press o f M IT and
W iley, N ew York.N Y (1954)
u Roussy, G. and Pearce, J.A., Foundations and Industrial Applications o f Microwave
and Radio Frequency Fields, W iley, N ew York. N Y (1995)
I’ Tian, Y ., “Practices o f Ultra-rapid Sintering o f Ceramics using Single Mode
Applicators". Cer. Trans., 2 1 .2 8 3 , (1991)
Ghaffariyan. S.R. and Methven. J.M .. “The Design o f a T M ,n Resonant Cavity
M icrowave Applicator as a Preheating and Crosslinking Die for Pultruded Composites".
M at. Res. Soc. Symp. Proc.,189.157. (1991)
31 Patterson. M .C .L.. Kimber. R .H . and Apte', P.S.. “The Properties o f Sintered Alumina
in a 2.45G H z Microwave Furnace", Mat. Res. Soc. Symp. Proc..189. 257. (1 9 9 1)
Jannev, M .A . and Kimrey. H .D ., "Diffusion-Controlled processes in Microwave-Fired
Oxide Ceramics", Mat. Res. Soc. Symp. Proc.. 189, 215. (1991)
Sweeney. M .P. and Johnson. D .L ., “Microwave Plasma Sintering o f Alumina". Cer.
Trans.. 21*. 365. (1991)
^ Patil. D.S. and Mutsuddy, B.C.. “ Microwave Sintering o f Alumina Ceramics in a
Single M ode Applicator", Cer. Trans.. 2 1 ,3 0 1 , (1991)
3; Swain. B., "Microwave Sintering o f Ceramics" ^ / v \ / a l p ro c .A 3 4 [3].76.( 1988)
34 Clark, D .E.. Folz, D .C ., Schultz, R .L ., Fathi, Z. andCozzi, A .D ., "Recent
Developments in the Microwave Processing o f Ceramics", M at. Res. Soc. B u ll... p41
(N ov. 1993)
37 Fang, Y „ Utilization o f Microwaves in Ceramic Processing. PhD dissertation. The
Pennsylvania State University, (1994)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
180
« Fang, Y ., “ Fabrication o f Porous Hydroxyapatite Ceramics by Microwave
Processing”, j
xfater. Res <1 [2], 490, (1992)
» Janney, M .A . and Kimrey, H .D ., “Microwave Sintering o f Alumina at 28G H z". Mat.
Res. Soc. Symp. Proc., 189, p ,(1 9 9 1 )
jo
Wroe, F.C .R . and Samuels, J„ “M icrowave Sintering o f Advanced Ceramics". Br.
Cer. Proc., 50. 3 9 ,(1 9 9 3 )
5' Katz, J.D. and Blake, R.D., “Microwave Sintering o f Multiple Alum ina and
Composite Components” , Qe r £ M//.. 70 [8], 1304, (1991)
Lewis, D .A ., "Microwave Processing o f Polymers-an Overview". M at. Res. Soc.
Symp. Proc., 2 6 9 ,2 1 , (1992)
” M ijo vic. J„ Fishbain. A. and W ijaya. J„ “Mechanistic Modeling o f Epoxy-Am ine
Kinetics. 2.Comparison o f Kinetics in Thermal and Microwave Fields".
Xfacrom oleculeS' 25. 986. (1992)
» W illert-Porada, M ., Krummel, T .. Rohde. B. and Moorman. D.. “Ceramic Powders
Generated by Metallorganic and Microwave Processing". Mat. Res. Soc. Symp. Proc..
2 6 9 .1 5 7 .(1 9 9 2 )
” W illert-Porada. M . and Vodegel. S., "Metallorganic and Microwave Processing o f
Ceramic Powders and Compacts". Cer. Trans.. 2 1.6 31 . (1991)
'<■ W illert-Porada. M ., Gerdes. T. and Vodegel, S., "Metallorganic and Microwave
Processing o f Cermets", Mat. Res. Soc. Symp. Proc., 2 6 9 ,2 0 5 . (1992)
” H aw ley. M .C . and Wei. J.. “Processing o f Polymer Composites in a M icrowave
Applicator", M at. Res. Soc. Symp. Proc.. 189 ,41 3 , (1991)
’» Mingos. D.M .P.and Braghurst, D .R ., "Applications o f Microwave Dielectric Heating
Effects to Synthetic Problems in Chemistry”, Chem. Soc. Rev., 20.1. (1991)
” W hittaker, A .G . and Mingos. D .M .P . “ The Applications o f M icrowave Heating to
Chemical Syntheses”, J. Micro. Pwr Elec. Eng., 29 [4], 195. (1994)
4° Mingos, D .M .P .. "Microwave Synthesis o f Inorganic Materials". ^ / v
5 [11
j.
8 5 7 ,(1 9 9 3 )
4> Mingos, D .M .P ., “Microwave Syntheses o f Materials and Their Precursors".
Chemical Processing o f Advanced Material, L.L. Hench and J.K. West (eds.), WileyPublishing, N ew York. N Y (1992)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
181
« Rao, K.J. and Ramesh, P.D., ‘‘Use o f Microwaves for the Synthesis and Processing o f
Materials". Bull. Mater. 5c/ ' 18 t4 ) - 447- ( 1995)
« Ahmad, I. and Clark, D.E., “Effect o f M icrowave Heating on Solid State Reactions o f
Ceramics", Cer. Trans., 21,6 05 , (1991)
« Rao, K.J. and Ramesh, P.D., “Use o f Microwaves for the Synthesis and Processing o f
Materials". B ull. M ater. S ch 18 l 4 J. 4 4 7, (1995)
4* Mathis. M .D ., Agrawal, D .K ., Roy, R, Plovnick, R. H. and Hutcheon. R..
“Microwave Synthesis o f Aluminum Titanate in A ir and Nitrogen". Cer. Trans.. 59.
5 5 7 ,(1 9 9 5 )
46 Mathis. M .D .. Agrawal. D.K ., Roy, R. and Plovnick, R. H.. “Microwave Synthesis o f
S iA lO N s” , Cer. Trans.. 59,533 , (1995)
47 Gedye, R „ Westaway, K. and Smith, F., “The Application o f Microwaves to Rapid
Chemical Synthesis”. Cer. Trans., 5 9 ,5 2 5 . (1995)
4« Ramesh. P.D.. Vaidhyanathan. B., Ganguli, M . and Rao. K.J.. “Synthesis o fP -S iC
Powder by use o f Microwave Radiation". /
\ f al
r l, s..
9 [12]. 3025. (1994)
4" Piluso. P.. Lequeux. N. and Boch. P.. “ Reaction and Densification o f M ullitc and
Aluminum Titanate Ceramics Heat-Treated in a 2 .45 G H z Microwave Furnace". j f}/}
C h i m . F r - ' I * 353-60. (1993)
» Boch, Ph., Lequeux, N. and Piluso. P.. “Reaction Sintering o f Ceramic Materials by
Microwave Heating”, Mat. Res. Soc. Symp. Proc.. 2 6 9 ,2 1 1 -1 6 (1992)
!' Cheng, J., Fang, Y.. Agrawal, R.F. and R. Roy. “M icrowave Synthesis o f LeadBarium Titanate”, Mat. Res. Soc Proc., 347, 513, (1994)
52 Binner. J.G.P.. Hassine. N .A . and Cross. T.E., "M icrowave Synthesis o f Non-Oxide
Powders”. Cer. Trans.. 59. 565. (1995)
53 Raner. K .D . and Strauss, C. R., “Influence o f Microwaves on the Rate o f
Esterification o f 2,4,6-Trimethylbenzoic Acid with 2-Propanol" / O re Chem- 57- 6231.
(1992)
*4 M eek, T .T . and Blake, R.D., “Ceramic-Ceramic Seals by Microwave Heating", j
Mat. Sci. L e tt - 5 ' 270. (1990)
55 Meek, T .T . and Blake. R.D.. U.S. Patent 4,606,748. (1986)
s* Kimrey, H. et.al., U.S. Patent 5,099,096 (1992)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
182
»’ Balbaa, I.S. andOda, S.J., Eur. Pat. Appl. no. 92301141.5, (1992)
» Oda, S.J., Balbaa, I.S. and Barber, B.T., “The Development o f New Microwave
Heating Applications at Ontario Hydro’s Research Division", Mat. Res. Soc. Symp.
Proc., 1 8 9 ,3 9 1 ,(1 9 9 1 )
« Arindam D .. Ahmad, I„ Whimey, E.D.and Clark, D.E., “Microwave (Hybrid) Heating
o f Alum ina at 2.45 G H z: 11. Effect o f Processing Variables, Heating Rates and Particle
Size”, Cer. Trans., 21, p329, (1991)
« Levin. E .M ., Robbins, C.R. and McMurdie, H.F., Phase Diagrams for Ceramists.
American Ceramic Society, Columbus, OH, (1964)
•i Thomas. H.A.J., Stevens, R„ Br. Ceram. Trans. J., 88. 144-151. (1989)
« Morosin. B.. Lynch, R .N .. “Structure Studies on AUTiO< at Room Temperature and at
6 0 0 °C '\ Acta C ry s t- B28, 1040, (1972)
« Kato, E... Daimon. K. and Kobayashi, Y . , j Ceram. Soc. J a p a n ' 81. (1978)
Lang, S.M .. Fillmore. C .L.. and Maxwell. L.H.. J. Res. Natl. Bur. Stds.. 48. 298.
(1952)
” Kameyama. T. and Yamaguchi. T . . j Cer. Soc. Japan'
^ 9 . (1976)
*> Kato, E.. Kobayashi. Y. and Daimon, K., //,/</.. 87, p81. (1979)
Asbrink. G. and M agneli. A .. Acta Chem. Scand- 21- 1278. (1967)
*' Bayer, G , , j i ess Com m on M etals»2*1.129, (1971)
« Ohya. Y ., Nakagawa, Z ., “Crack Healing and Bending Strength o f Aluminum
Titanate Ceramics at High Temperature”, j Am Cer. S o c- 71 [5], C232. (1988)
70 Parker. F.J., Rice, R .W ., “Correlation Between Grain Size and Thermal Expansion
for Aluminum Titanate Materials”, j Am Cer. Soc- 72 [12], 2364. (1989)
*i Cleveland, J.J.. Brandt, R .C . , /
Cer. Soc- 61- 420, (1973)
^ Thomas, H.A.J.. Stevens, R., "Aluminum Titanate- A Literature Review Part 2:
Engineering Properties and Thermal Stability”, Br. Ceram. Trans. J.. 88. 184, (1989)
” Freudenberg, B. and M ocellin, A ., "Aluminum Titanate Formation by Solid-State
Reaction o f Coarse A120 3 and T iO : Powders”, / Am Cer. S oc- 71 [*]• 22, (1988)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
183
’«Freudenberg, B. and M ocellin. A ., “ Alum inum Titanate Formation by Solid-State
Reaction o f Fine A U O j and T iO : Powders", j
A m Cer. go c „ 70 [1 ]. 33. (1987)
Lee, S. C. and Wahlbeck. P. G., "Chemical Kinetics and Mechanism for the
Formation Reaction o f A I2T i 0 5 from Alumina and Titania”, /y/g/, f emp S c i•• 21. 27.
(1986)
Prasadarao, A .V ., et al., “Enhanced Densification by Seeding o f Sol-Gel-Derived
Aluminum Titanate". j
Am Cer. Soc- 75 [6]- 1529, (1992)
” Woignier. T ., et al., “A l20 3- T i0 2 and A l2TiO< Ceramic Materials by the Sol-Gel
Process", j
\ ’on-Cryst. S o lid s•
325. (1988)
'i DeVries, R.C and Roy. R., “A Phase Diagram for System T i-T iO : Constructed from
Data in the Literature", A m Qer. S o c .B u ll' 33 l l 2], 370, (1954)
^ Kofstad. P., Nonstoichiometry, Diffusion and Electrical Conductivity in Binary' Metal
Oxides, R.E. Krieger Publishing Co. Malabar, F L (1983)
«° Porter, V .R .. “Studies in the Titanium-Oxygen System and the Defect Nature o f
Rutile". PhD dissertation, the Pennsylvania State University (1965)
Hutcheon. R. M .. et al.. j
xiicro . Power Elec. Energy" 27 [2|. 93-102 (1992)
Hutcheon. R. M .. et al., “RF and Microwave Dielectric Measurements to 1400°C and
Dielectric Loss Mechanisms”. Mat. Res. Soc. Symp. Proc.. 269. 541. (1992)
« Rase, D.E. and Roy, R.. “Phase Equilibria in the System BaO-TiO;" j Am C er Soc38 [3], 102.(1955)
» Kirby, K .W ., Wechsler, B .A . , j Am , Cer. S oc•< 74 [8], 1841. (1991)
« Negas. T ., et al.. “B aT i4O y Ba2T i9O 20-Based Ceramics Resurrected for M odem
Microwave Applications”. Cer. B u ll•* 7*2 [11.80. (1993)
Jaffe. B., Cook, W .R. and Jaffe. H., Piezoelectric Ceramics, R.A.N. Publishing.
Marietta, O H (1971)
«■ M uller, O. and Roy, R .. The M ajor Ternary Structural Families. Springer-Verlag.
New York, N Y (1974)
«» Dickson, J.G.. Katz, W . and Ward. R.. "Compounds with the Hexagonal Barium
Titanate Structure", j A m Chem. Soc■' 83 - 3026, (1961)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
184
« Cross, L.E. and Hardtl, K.H.. “Ferroelectrics”, reprinted from Encyclopedia o f
Chemical Technology, 10,1-30, J. W iley & Sons Inc. (1980)
« Newnham, R.E., Structurc-Property Relations, Springer-Verlag, N ew York. N Y
(1975)
■>' Vivekanandan, R. and Tutty, T.R .N ., “Characterization o f Barium Titanate Fine
Powders Formed from Hydrothermal Crystallization”. P o W e r Tech ' 57. 181. (1989)
« Hennings, D. and Schrcinemacher, S., “Characterization o f Hydrothermal Barium
Titanate”, . / £ u r Ceram. Soc- 9 <41,(19 9 2)
11 Hennings, D „ Rosenstein G. and Schreinemacher, H .. j
£ u r Ceram Soc•> 8.107.
(1991)
M Fang, T., Lin. H. and Hwang, J., “Thermal Analysis o f Precursors o f Barium Titanate
Prepared by Coprecipation".y Am Cer. Soc-' 73 [11], 3363, (1990)
•» Beauger. A ., M utin, J.C. and Niepce. J.C.. “Synthesis Reaction of Metatitanate
B aTiO j: Part 1; Effect o f Gaseous Atmosphere Upon the Thermal Evolution o f the
System B a C 0 3-T iO :", j
Mat. s cj „ 18. 3041. (1983)
*• Beauger. A ., M utin. J.C. and Niepce. J.C.. "Synthesis Reaction o f Metatitanate
B a T i0 3: Part 2; Study o f Solid-Solid Reaction Interfaces", j
Mat.
18. 3543. (1983)
Templeton. L.K.. and Pask, J.A., “Fonnation o f B a T i0 3 From B a C 0 3 and T iO ; in A ir
and in C O :” . j
Am Cer. Soc- 42. 212. (1959)
’« Suvama, Y .. Oda. Y . and Rato, A., chem. L e tt-
^87. (1979)
« C oum il. M .. Soustelle, M. and Thomas, G..
M e t- ^3. 77. (1979)
q xi^
Schmutler, H.J., Antony, M .M . and Sandhage, K .H ., “A Novel Reaction Path to
B a T i0 3 by the Oxidation o f a Solid Metallic Precursor”, j Am Cer S o c- 77 [3], 721.
(1994)
mi Katayama, S. and Sekine, M ., “Low-Temperature Synthesis o f B a(M g|/3Ta3/3) 0 3
Perovskite by the Metal Alkoxide Method”, j
Mater. C hem - 2 [8]. 889. (1992)
m: Guo, R., Bhalla, A.S. and Cross, L.E., “B a(M g|/3Ta2/3) 0 3 Single Crystal Fiber Grown
by the Laser Heated Pedestal Growth Technique”, j
Ap p [ ph\-s- 75 [9]. 4704. (1994)
mj Guo, R.. Bhalla, A.S., Roy, R. and Cross. L.E., “Candidate HTSC Film Substrates o f
Complex oxide Perovskite Compositions”, Mat. Res. Soc. Symp. Proc., 341. 215.
(1994)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
185
»» Furuva, M . and O chi.A „ “Microwave Dielectric Properties for B a f M g ^ T a ^ K V
A fM g i/iW i/iJ O j , jp r i' J Appl. Phys- 33 [9B ], 5482, (1994)
Kagata. H. and Kato, J., “Dielectric Properties o f Ca-Based Complex Perovskite at
Microwave Frequencies",^,,. j Appl. Phys * 33 [9B ], 5463, (1994)
i<* Nagai, T „ Inuzuka, T. and Sugiyama, M ., “Contribution to Dielectric Constant to
Change in Temperature Coefficient o f Resonant Frequency in (Ba|.xSrx) ( M g ,3T a :/3) 0 ,
Compounds". j p n j
Ap pl. Phys- 31 [9B], 3132, (1992)
'»• Ravichandran, D. et al., “Sol-Gel Synthesis o f B a(M g |/3Ta:/3) 0 3: Phase Pure Powder
and Thin Films". \ ( a( g es B u ll- 31 [71. 817, (1996)
'0* Katayama. S., et al.. “Low-Temperature Synthesis o fB a (M g |,3Ta: 3 ) 0 3 Ceramics
from B a-M g-T a Alkoxide Precursor", j Am. Cer. S o c- 79 [81- 2059. (1996)
'« Jack. K .H . and Wilson. W .I., "Ceramics Based on the S i-A l-O -N and Related
Systems". S u tu re Phys. S c i- 238<28- O 972)
Oyama. Y . and Kamigaita. 0 .. “Solid Solubility o f Some Oxides in Si3N 4". j p n j
Appl. P h ys- 10.1637.(1971)
"I Jack. K. H .. “ Review: SiAlO N and Related Nitrogen Ceramics", j
\[ a l S c i- 11-
1135.(1976)
"j Ekstrom. T. and Nygren. M ., “S iA lO N Ceramics”, j
Am Cer S oc- 73 l 2 l- 2- 9
(1992)
in Hwang, S. and Chen, I., “Reaction Hot Pressing o f a '- and P'-SiA lO N Ceramics", j
Am. Cer. S o c - 77 [1], 165 (1994)
n4 Ohashi. M . et. al., “Solid Solubility o f Alum inum in O '-S iA lO N ". j
C er Soc-
76 [8], 2112, (1993)
115 Oyama. Y ., Yogyo-Kyokai-S hi- 82-
(1974)
n<> Lejus A .M ., p ev tfautes Temp, et R e fra c t- 1’ 53, (1964)
'>■ Oauckler, L.J., Lukas, H .L. and Petzow, G .. “Contribution to the Phase Diagram
S ijN r A lN -A L O j-S iO i" . j Am. Cer. S o c- 38* 3^6, (1975)
m Bower. R .M ., Carr, A.J. and Hendry, A ., “ Fabrication and Microstructure o f
S iA lO N -S iC Composites". Br. Ceram. Proc., 42. 25, (1989)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18 6
Edrees, H.J., Holling, G.E. and Hendry, A., “ S iA lO N Bonding Systems for Silicon
Carbide Composites”, Br. Ceram. Trans., 94 [2], 52, (1995)
im
Henshall, J.L., N G , S.G. and Wakeman, R.J., “Synthesis and Microstructure o f
SiA lO N Membrane Filters” , Br. Ceram. Proc., 53. 59, (1994)
i.'i Lee, J. G. and Cutler, I. B., Ceram. B u ll '
[9J, 869. (1979)
i-’-’ Cho, Y . W . and Charles, J. A., “Synthesis o f Nitrogen Ceramic Powders by
Carbothermal Reduction and Nitridation. Part 2: Silicon Aluminum Oxynitride
(S iA lO N )” . M a te ria ls Sci. Tech- 7’ 399, (1991)
I2J Higgins, I. and Hendry. A ., Br. Ceram. Trans. J " 8 5 ,1 6 1 , (1986)
i:i Sugahara, Y ., Kuroda, K. and Kato. C., "The Carbothermal Reduction Process o f a
Montmorillonite-Polyacrylonitrile Intercalation Compound”, j
\{a( $cj - 2 3 . 3572.
(1988)
|23 Dijen. F. K.and Metselaar, R., "Reaction-Rate-Limiting Steps in Carbothermal
Reduction Processes”, j
£ er Boc-, 88 [1], pp 16. (1985)
i2» Shiosaki. T. and Kawabata. A., “Future Trends in Piezoelectric Materials and
Applications". Ferroelectrics. 95. 9. (1989)
|2' Ounaies, Z.. Selmi. F.. Varadan. V .K .. Varadan. V .V . and Megherhi. M .. “M icrow ave
Calcination o f Conventionally and Sol-Gel Prepared Lead Zirconate Titanate”. \ f „ ,
Le({.. 17. 13. (1993)
i2» Sugihara, S., “Sintering o f Piezoelectric Ceramics with CO-. Laser", j w , j
jp p (
Phy5., 31 [9B ], 3037, (1992)
|2’ A lliouat, M .. M azo, M ., Desgardin, G. and Reveau. B., “Microwave Sintering o f
Lead-Based Perovskite Ceramics for Capacitors". M at. Lett ' 5 [9], 328. (1987)
no Harrison, W .B ., Hanson, M .R .B . and Koepke, B.G., Mat. Res. Soc. Symp. Proc..124.
2 7 9 .(1 9 8 8 )
131 Komameni. S., Li. Q.. Stefansson, K .M . and Roy. R.. “Microwave-Hydrothermal
Processing for Synthesis o f Electroceramic Powders”, j
M at. Res ' 8 [ 12]. 3176. (19 93 )
132 Budd, K .D .. Dey, S.K. and Payne. D .A ., Br. Ceram. Proc.. 3 6 .1 0 7 . (1985)
133 Chen, K .C . and MacKenzie, J.D., “Crystallization Kinetics o f Metallo-Organics
Derived P Z T Thin Film ”, M at. Res. Soc. Symp. Proc., 180.663, (1990)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
187
'» Gupta. S .M . and Kulkami, A.R., “Role o f Excess PbO on the Microstructure and
Dielectric Properties o f Lead Magnesium Niobate". j
\ f a t Res„ 10 [4], (1995)
'» Selby. J., “ Production and Uses o f Zircon”, p \y p j . , 31 [12], 7. (1991)
u* Itoh, T ., “ Formation o f Polycrystalline Zircon (Z rS iQ ,) from Amorphous Silica and
Amorphous Z irc o n ia".,/ Cryst. G ro w th *25, 223, (1992)
m Curtis, C.E. and Sowman. H.G.. “Investigation o f the Thermal Dissociation.
Reassociation, and Synthesis o f Zircon” ,/ Am Cer. Soc-' 36,190. (1953)
•» Kanno, Y.. "Thermodynamic and Crystallographic Discussion o f the Formation and
Dissociation o f Zircon”, j
13,1 Kakamu, Y .. j
\ ( a l Sci•• 2 4 .2 4 1 5 , (1989)
Refract. Japan' 39 [9], 510, (1987)
i«> Bayer, G., "Thermal Expansion Anisotropy o f Oxide Compounds". Proc. Br. Ceram.
Soc., 22, 3 9 ,(1 9 7 2 )
i4i Subbarao, E.C.. Agrawal, D.K., McKinstry, H .A .. Sallese. C.W . and Roy. R..
“Thermal Expansion o f Compounds o f Zircon Structure".,/ Am C er Soc - 23 [5],
1246.(1990)
n: Thorpe, M .L . and Wilks. P.H.. "Electric-Arc Furnace Turns Zircon Sand to Zirconia".
Chem. Eng ' 28 [26]. 117, (1971)
143 Prasad. S.V. and Rohatgi, P.K.. \ [ a( gci a n d Eng ' 8 0 ,2 1 3 . (1986)
144 Barrie, J.D. and Meshishnek, M.J., Mat. Res. Soc. Symp. Proc.. 180.347. (1990)
'45 M illet. P. and Hwang, T., "Preparation o f T iB : and Z rB 2. Influence o f a
Mechanochemical Treatment on the Borothermic Reduction o f Titania and Z irco nia".,/
Mat. S c i"
351, (1996)
'46 Sundaram. V .. Logan. K.V. and Speyer, R.F.. “ Reaction Path in the Formation o f
Titanium Diboride by a Magnesium Thermite Process", Cer. Eng. Sci. Proc.. 56. I l l
(1995)
147 Calka, A. and Radlinski, A.P., “Formation o f T iB : by Mechanical Alloying", j
le s s
Common M eta ls' * ^ L L23, (1990)
i4« Radev, D .D ., and Klisurski, D., "properties o f T iB 2 Powders Obtained in a
Mechanochemical W ay”, j Auoys and C om pounds' 206, 39, (1994)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
188
' « Jiang, Z ., and Rhine, W .E., "Preparation o f Titanium Diboride from Titanium
Alkoxides and Boron Powder", Chem. X faier- 4. 497, (1992)
Annual M inerals Review. Cer. B u ll-
(1994)
m M iller, R .L ., et al., "Atmosphere Effects on Combustion Synthesis o f SiC and T iB ;
Powders". Cer. Eng. Sci. Proc., 56. 77. (1995)
is: Watt. G .L .. et al., "Consolidation o f T iB : Powders Produced By Plasma Synthesis o f
Cheap Raw Materials". Cer. Eng. Sci. Proc.. 5 6 .1 0 3 , (1995)
hi Janney, M .A ., "Mechanical Properties and Oxidation Behavior o f a Hot-pressed SiC15-vol% -TiB 2 Composite". Cer. B u ll ' ^
[-1- 322, (1987)
i« Garard. R.S., Fathi, Z. and Wei. J.B.. "Materials Processing via Variable Frequency
M icrowave Irradiation". Cer. Trans-' 59. 117. (1995)
hi Fathi. Z. et al.. "Processing and Characterization o f a Polymer Matrix Composite
Using Variable Frequency Microwave Heating", Cer. Trans - 59. 4 4 1 .(1 9 9 5 )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
VITA
M ilton Mathis was bom in Savannah, Georgia (U S A ) to Enoch and Yvonne Mathis
in 1961.
He was a graduate o f Alfred E. Beach High school in 1979. after attending
several elementary schools.
In 1984 he received a Bachelors o f Science degree in
Chemistry at Savannah State College in Savannah. Georgia.
In 1987 he received a
Masters o f Science degree in Chemistry at Tuskegee University in Tuskegee. Alabama.
He worked as a Research Instructor at Tuskegee University following completion o f the
Masters degree.
From 1988 until 1990 he worked at the 3 M company in St. Paul.
Minnesota on several research projects.
In the fall o f 1990 he joined the graduate
program in materials at the Pennsylvania State University for the degree o f Doctor o f
Philosophy. During these years he has worked on several microwave research projects at
Penn State and 3 M .
He has authored several papers in the open literature and technical
reports that are 3 M restricted.
Milton is a member o f the American Ceramic Society.
American Chemical Society. Sigma X I.
Materials
Research Society.
International
Microwave Power Institute. 3M Technical Forum and American Society o f Metallurgists.
His publications include:
Mathis, M .D ., Agrawal. D .K ., Roy, R. and Plovnick, R. H.. "Microwave Synthesis o f
SiAlO N s". Cer. Trans.. 5 9.5 3 3 .(19 95 )
Mathis. M .D ., Agrawal, D .K ., Roy, R, Plovnick, R. H. and Hutcheon. R., "M icrow ave
Synthesis o f Aluminum Titanate in A ir and Nitrogen". Cer. Trans.. 59, 557, (1995)
Mathis, M .D .. Dewan, H.S.. Agrawal, D .K ., Roy, R, and Plovnick. R. H.. “M icrowave
Processing o f N i- A I; 0 3 Composites", Cer. Trans., 3 6 ,4 3 1 . (1993)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 881 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа