A STUDY ON THE CRYSTAL GROWTH OF SELECT II-VI OXIDES BY CZOCHRALSKI AND BRIDGMAN TECHNIQUES By JALAL MOHAMMAD NAWASH A dissertation submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY WASHINGTON STATE UNIVERSITY School of Mechanical and Materials Engineering DECEMBER 2006
226
Embed
A STUDY ON THE CRYSTAL GROWTH OF SOME II-VI … · a study on the crystal growth of select ii-vi oxides by czochralski and bridgman techniques by ... 3.3.4.4 more optical properties
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
A STUDY ON THE CRYSTAL GROWTH OF SELECT II-VI OXIDES BY
CZOCHRALSKI AND BRIDGMAN TECHNIQUES
By
JALAL MOHAMMAD NAWASH
A dissertation submitted in partial fulfillment of
the requirements for the degree of
DOCTOR OF PHILOSOPHY
WASHINGTON STATE UNIVERSITY
School of Mechanical and Materials Engineering
DECEMBER 2006
To the Faculty of Washington State University:
The members of the Committee appointed to examine the dissertation of
JALAL MOHAMMAD NAWASH find it satisfactory and recommend that it be accepted.
Chair
ii
ACKNOWLEDGMENTS
I would like to thank Dr. Kelvin Lynn whose support and guidance throughout the entire period
of this research made this work achievable. Several materials’ characteristics were carried out in several
laboratories at both Washington State University and University of Idaho. Many thanks go to the
collaborators of these laboratories especially Dr. Roger Willett, Dr. Brendan Twamley, and Scott
Cornelius. I am also grateful for every employee, graduate students, and undergraduate students who
contributed to this work. I thank Robert Novotney and Lloyd Pilant for their technical assistance, Guido
Ciampi, Kelly Jones, Romit Dhar, Charles Shawley, Christie Skrip, and Russ Tjossem either for helping
in measurements or for sharing their ideas.
This research was sponsored by: Space Missile Defense Command (SMDC). Contract
Number: DASG60-02-C-0084 and VLOC Incorporated.
iii
A STUDY ON THE CRYSTAL GROWTH OF SELECT II-VI OXIDES BY
CZOCHRALSKI AND BRIDGMAN TECHNIQUES
Abstract
By Jalal Mohammad Nawash, Ph.D.
Washington State University
December 2006
Chair: Kelvin G. Lynn
The crystal growth of ZnO-TeO2 system was experimented by Czochralski and Bridgman
techniques. The series of many runs and experimentations helped optimize the growth process,
which was faced by a lot of difficulties. These difficulties include, but are not limited to, the
evaporation of TeO2 material above 700 ºC, the formation of more than one phase during the
growth, and the lack of a ZnO-TeO2 single crystal to start the growth. It was concluded that the
main and most persisting problem is that there is no stable phase, in the system that forms a line
component at which the crystal growth should be attempted. However, Zn2Te3O8 and ZnTeO3
single crystals were grown using Czochralski and Bridgman techniques, respectively. It was
possible to study some of their important optical and electrical properties for the first time.
The phase diagram of this system was investigated using powder x-ray diffraction and
scanning electron microprobe. CrystalDiffract 1.3 for Windows software was used to simulate x-
ray patterns to find the percentages of the resulting phases. It was found that the type of forming
phases might be affected by the process, whether if it was calcining, melting, or pulling.
Moreover, the history of the material plays an important role in determining what phases form.
iv
The glass form of ZnO-TeO2 system was studied as well for this research. One important finding
is that the cut-off band edge of this glass depends greatly on the thickness of the sample used.
Dielectric constants and resistivities of several glasses were determined.
Bridgman technique was used to grow CdTe2O5 single crystals. These crystals are
transparent to visible light, and have a mica-like structure. Optical and electrical properties of
these crystals, like the dielectric constant and resistivity, of these crystals, were investigated.
v
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ………………………..…………………………………………….iii
ABSTRACT…………………………………………………………………………………….iv
LIST OF TABLES……………………………………………………………………………...x
LIST OF FIGURES…………………………………………………………………………….xiii
CHAPTER
1. BACKGROUND AND LITRATURE REVIEW....…………………...........…....................1
5.40 X-ray for 25:75 glass....................................................................................................156
5.41 Phase diagram for the ZnO-TeO2 system, it shows the glass forming region. The solid
line corresponds to cooling rates of 1 K/min and the dotted line corresponds to cooling
rates of 10 K/s. The phase diagram was taken from Bürger et al [10]............................157
5.42 Transmission curves for the glasses shown in Figure 5.36...........................................158
5.43 Cut-off edge as a function of thickness in mm for the 32:68 glass...............................159
5.44 Dependence of the transmission on the angle of the glass sample...............................160
5.45 Absorption curves for both the 35.5:64.5 glass and Zn2Te3O8 single crystal...............161
xxiv
DEDICATION
To my late mother and my late father,
for their unconditional love and support, may God bless their souls.
xxv
CHAPTER ONE
BACKGROUND AND LITERATURE REVIEW
1.1 Introduction
The II-VI oxides have been the focus of many studies for there useful optical and electronic
properties. In this research, Czochralski (CZ) as well as vertical Bridgman techniques were used
in an effort to grow single crystals of Zn2Te3O8 and ZnTeO3. There is no report in literature of
any attempt to grow these crystals using CZ or Bridgman techniques, yet, very small crystals
were produced hydrothermally to study the crystal structure and other simple physical properties
like color, appearance, and density.
The growth of Zn2Te3O8 and ZnTeO3 was faced with challenging difficulties. The phase
diagram of ZnO-TeO2 system does not show the formation of stoichiometric compounds for
either of these two materials. Attempts to grow these crystals were carried out to replicate other
researchers’ efforts that effectively grew nonstoichiometric oxide crystals.
Phase diagram of the ZnO-TeO2 system was investigated by calcining, melting, and pulling
mixed powders of various mole percentages. The primary focus was on 21%:79% - ZnO:TeO2
by mole. The glass forming ability of this system was also examined for this mole percentage.
Different mole percentages of ZnO:TeO2 glasses were also formed and their transmission
properties were investigated.
On the other hand, another II-VI oxide, CdTe2O5 was successfully grown using top cooling
vertical Bridgman technique. Earlier researchers used only the CZ method. Electrical and optical
properties of this crystal were compared to the properties of the crystals grown by CZ technique.
For the three single crystals that were grown, optical band gap, resistivity, dielectric constant and
other properties were determined.
1.2 Literature Review
Although the development of crystal growth started early in the twentieth century [1],
Czochralski (CZ) crystal growth was well established by the mid-1950s. It had shown a great
potential to pull oxide crystals [2,3,4], as well as semiconductor crystals such as silicon [5] and
germanium [6]. Many other types of crystals were also grown by CZ technique [7,8,9].
To grow a crystal using CZ crystal growth technique, the material has to be melted in a
suitable crucible, and then a seed is lowered onto the surface of the melt, such that the clean
surface of the seed touches the surface of the melt. Then the rotating seed is pulled slowly to
form the crystal. A schematic diagram is shown in Figure 1.1.
Figure 1.1 Schematic diagram of CZ setup. Source after modification:
http://rcswww.urz.tu-dresden.de/~cwinkler/poverview.htm accessed on August 24 2006.
2
The material is placed inside a suitable crucible and heated by a radio frequency (RF) coil
[10] or a regular ceramic heater. The mixture of the materials is preferred to be at the congrue
melting point of the constituents to avoid complications of forming undesired phases while
growing, but some researchers were able to pull single crystals at incongruent points [11]. Ot
workers reported the growth of single crystals from non stoichiometric melts [12, 13] and others
grew multiphase semiconductor crystals at the peritectic phase transformation [14].
Several parameters have to be controlled during the crystal growth. . These parameters
include the temperature gradient, the melt-crystal interface shape, the rotation of the seed, th
pulling rate, and the growth direction. Because of the dynamicity of CZ crystal growth, some o
these parameters have to be modified as the growth process progresses. Since CZ growth is very
sensitive to these parameters, the outcome could be different from one researcher to another.
Different outcome
nt
her
e
f
s can also occur for the same operator where two successive runs yield
evel
al, the melting temperature of
different results, even if both of the runs have almost the same constraints. The temperature
gradient across both the melt and the space above the melt, which will in turn, affects the axial
temperature gradient on the seed and the growing crystal is the most important variable that
needs to be controlled. A good axial temperature gradient above the melt surface will help grow
a crystal with the least amounts of defects. Defects that can occur include macroscopic and
microscopic cracks, high intensity of dislocations, impurities and/or dopant inhomogeneities,
core and/or surface facets and other defects [15]. Another parameter that will determine whether
defects occur is the dopant percentage level. Cracks often appear if the dopant percentage l
is incorrect [16].
The temperature gradient for both the melt and the room above of it strongly depends on the
size and the emissivity of the crucible and the melt, the melt materi
3
the material, the size of the crystal needed to be grown and its emissivity, as well as the melt’s
emissivity [17]. Temperature gradient also depends on the diameter and height of the build up
around the growing crystal, the type of insulation material used around the crucible and the
growing crystal. A good axial and radial temperature gradient in the melt will maintain suitable
convection currents that are needed to stir the melt. These convection currents will help conduc
the heat from the hot spots in the crucible to the cold ones and prevent severe temperature
gradients from existing. In the case where the melt contains dopants, these convection curr
will keep the concentrat
t
ents
ion of these dopants uniform through out the entire volume of the melt,
al is considerably less that that for the
e
s done, either by polishing or cutting.
which is a superior benefit of CZ technique over other techniques.
The melt-crystal interface shape is strongly affected by the internal radiative heat transfer
[18,19]. It was found that a deflected interface towards the melt is promoted by heat transfer. In
oxides, this happens because the absorptivity in the cryst
m lt. Also, because of that effect, it was found that a steady spoke line pattern can be seen when
the melt is not assumed to be transparent, in this case the Marangoni effect is enhanced due to
the existence of a thermal gradient [20]. Many simulations were performed to better understand
the convection flow of the melt and its effect on both the melt-crystal interface and spoke
patterns [21,22,23,24].
Increasing the rotation of the seed/crystal could result in changing the solid-melt interface
shape [2,16] from convex to flat. This has the good effect of displacing the facets from the center
of the crystal to the sides. This will make it an uncomplicated process to get rid of these facets
when the growth i
Verifying a suitable thermal gradient in the melt mostly depends on noticing the spoke line
patterns that appear on the surface of the melt [25]. In most melts, these spoke lines form a star-
4
like shape, in which the center of the star is almost at the center of the crucible and the branches
of the star rotate around the center in a slow motion that ranges from 1-3 rpm. This speed
depends greatly on the material of the melt, the diameter and the height of the crucible and other
es
t
thermal gradient becomes larger, and this could cause the
p,
placed
the
variables. Spoke lines form a star-like shape at the center of the crucible in most melts, in TeO2-
CdO melts, spoke pattern center is shifted towards the crucible edge and has very small branch
that move slowly, along each line, towards that center, while new branches appear at the tip of
each spoke line.
Another variable that is essential to the crystal growth is the pulling rate [26, 27]. In mos
materials, pulling rates range from 0.5 to 20 mm/hr. The pulling rate should be adjusted as the
crystal grows and gets bigger, since the heat transfer dynamics change accordingly. For example,
as the diameter of the crystal increases, more melt is crystallizing in a shorter time, and this will
make the latent heat released bigger and cause the flow dynamics in the melt to change.
As the crystal gets longer, the axial
crystal to crack [28]. A good, but not adequate, solution to this problem is to slow the pulling
rate to its minimum value. Some researchers use a heat shield and/or an afterheater [28,29] to
reduce the axial thermal gradient. Another problem that could rise in growing big crystals is the
melt level dropping in the crucible; again this changes the fluid dynamics and heat transfer.
Some researchers [30] overcame this difficulty by melting the material in a two- crucible setu
such that one of them is inside the other, a powder supply system provides the outer crucible
with a powder to compensate for the loss of melt level due to growth. Other researchers
crucible assembly on top of a stepping motor that slowly raises the setup, which holds the
crucible as the crystal grows [31, 32].
5
Growth direction is another variable that could be significant to crystal growth [27, 33],
Some growth directions are easier to execute than others are. Some crystals have the likelihood
to develop certain types of defects in one direction, but when grown in another direction, there
less probability for them to appear. The crystal growth direction depends on the seed used,
seed is in a certain direction, then the growth will be in the same direction as the seed. Some
researchers use seeds that were made out by slow spontaneous nucleation of the melt [34] or
used seeds that were made with the help of a platinum wire [35] or iridium wire [36].
In the Bridgman technique, a thermal gradient is utilized to grow large single crystals either
by lowering the melted charge through a hot zone with a thermal gradient or by creating the
gradient via electronic control. Better crystals are grown when using the electronic control
gradient, since moving the melt down the thermal gradient zone may disturb the formation of
defect free single crystal. In the mid twenties of the past century, Bridgman was
is
if the
a
able to grow
me
ell
rmal gradient. Some crystals that are not well grown by CZ technique can be
gr
tallic crystals several times. A modified version of Bridgman technique was introduced by
Stockbarger's method [37], in which the thermal gradient was made steeper to grow large
crystals of Lithium Fluoride by separating the hot zone from the cold zone via a partition made
of platinum. To grow a crystal that has the same direction as the seed, the crucible may have a
thin vertical hollow extension at the bottom where a cylindrical seed can fit [38]. More
sophisticated furnaces were found when the multizone furnace was introduced. In this furnace,
which was first built by Mellen [39] the axial temperature was controlled via local heaters
separated by insulation material stacked vertically. This provided a uniform, short, and w
maintained the
own by Bridgman method; some of these crystals are those of volatile melts.
6
The solid-liquid interface shape in a Bridgman-Stockbarger furnace is generally convex in th
hot zone (solid point of view), but reverses to concave in the cold zone [40]. This mainly
happens as a result of heat transfer between the crucible and the heat source which leads to
curving of the isotherms [41]. The study of the interface shape was the focus of many authors for
its importance in determining the quality of the grown crystal [42,43]. For example, it is known
that defects tend to spre
e
the
ad normal to the growth interface, if the melt was concave (melt point of
ll form inside the crystal, but if the melt was convex, defects form on the
out
id
or
,
expected to be
e used in
view), defects wi
side. Flat interfaces are best for melts with dopants, since it provides uniform radial
distribution.
Some Bridgman techniques use a horizontal thermal gradient, in which the polycrystalline
material is placed in a boat, and this boat is exposed to a uniform thermal gradient [44]. In this
process, which is more complicated than vertical Bridgman is, at least 40% of the solid-liqu
interface is free of contact with the crucible. This causes no chemical, mechanical, thermal
kinetic interactions occur. This provides the situation of growing defect (dislocation) free
crystals [45]. Finally, Bridgman method is used mostly to grow x-ray and gamma ray detectors
II-VI and III-V semiconductors, and piezoelectric materials, in addition to the growth of some
metal single crystals.
1.3 Motivation
Materials with wide band gaps are used in optoelectronic devices, as well as in acousto-
optical instruments. Good quality grown Zn2Te3O8 and ZnTeO3 crystals are
transparent to visible light. This means that it has a wide band gap, with all the benefits that
come with this property. As soon as these crystals are successfully grown, they could b
the solar cell industries. The grown crystal has a resistivity of the order of 1013 ohm.cm or
7
higher, once doped with the appropriate dopants before the growth; it can be used in many
applications that involve semiconductor manufacturing. If the resultant crystal has low ligh
absorption and high transmissivity, this makes it a candidate to become a laser crystal. This is
conditioned by the fact that the population inversion can be achieved [46]. It has been reported
that the II-VI oxide crystals have very high refractive indices and are optically active. They
present non-linear optical propertie
t
s [47, 48], second harmonic generation (SHG) effect, and
their useful properties. When the
ing
oxide (ZnO) crystals have a wide band gap width of 3.3 eV at room temperature
hort wavelength lasers and light emitting diodes (LED) [55]. The average refractive index and
e average static dielectric constant of ZnO crystals are 2.0 and 10.0, respectively [56]. On the
ls have useful applications in acousto-optic devices
birefringence. This makes it a good material in manufacturing fiber optics, polarizers, wave
plates, depolarizers, and optical filters and many other optical instruments.
The study of this system as a crystal was not examined in the literature. A quick investigation
of the crystal structure and phase formation was done by some authors [49,50,51,52].
Unfortunately, these studies came because these authors were studying the glass that forms when
this material is quenched from melt, not because they were interested in the crystal form of this
system. Similar II-VI oxides have attracted attention for
powders of CdO and TeO2 are mixed in certain mole percentages and melted for growth by CZ
technique, the resulting crystal is transparent with piezoelectric properties [48]. Other mixing
percentages, which produce MTeO3 crystals, where M stands for Zn or Cd, show promis
nonlinear optical properties [53].
The zinc
[54]. This makes the crystal a good candidate for applications in optoelectronic devices such as
s
th
other hand, Paratellurite (TeO2) crysta
8
[57,58], many of which are used in data display devices (DDD)[59]. TeO2 crystal has band gap
In addition to the benefits that II-VI oxide crystals may have to offer, glasses of this group,
ns in SHG after thermal poling [61, 62], and
] H. J. Scheel, The Development of crystal growth technology: Crystal growth technology
of 3.5 eV [60] and a refractive index of about 2.2 [56].
particularly the glass of ZnO- TeO2 have applicatio
can show non-linear optical properties. ZnO-TeO2 glasses are also known for their high
refractive indices, and dielectric constants [63], and they are also chemically stable [64].
References
[1 .
Czochralski growth of oxides, Journal of crystal growth 264: 593 - 604, 2004.
] L. Fornasiero, E. Mix, V. Peters, G. Petermann, and G. Huber, Crest. Res. Technol., 34, No.
] N.V. Abrosimov, H. Riemann, W. Schröder, H.-J Pohl, A. K. Kaliteevski, O. N. Godisov, V.
allmoos, P. Bastie, B. Hamelin, and R. K. Smither, Growth and properties of Ge1-xSix osaic single crystals for γ-ray lens application, Journal of Crystal Growth, 278: e495 - e500.
S. X. Gao, W. S. Zhan, Y. X. Li, J. P. Qu, and G. Z. Xu, ingle crystal growth and phase diagram of Tb2Fe17 compound, Journal of Crystal Growth, 222
0] L. C. F. Blackman, p. H. Dundas, A. W. Moore, and A. R. Ubbelohde, An induction furnace
John Wiley & Sons, Ltd. New York, 3 - 14, 2003. [2] C. D. Brandle, [3] S. Kumaragurubaran, D. Krishnamurthy, C. Subramanian, and P. Ramasamy, Journal of crystal growth, 211: 276 - 280, 2000. [42: 255 - 260, 1999. [5A. Korolyov, and A. J. Zhilnikov , 29Si and 30Si single crystal growth by mini-Czochralski technique, Crys. Res. Technol., 38 No. 7-8: 654 - 658, 2003. [6] P. Dold, Czochralski growth of doped germanium with an applied rotating magnetic field, Crys. Res. Technol., 38 No. 7-8: 659 - 668, 2003. [7] K. Sankaranarayanan, and P. Ramasamy, Journal of crystal growth, 193: 552 - 256, 1998. [8] N.V. Abrosimov, A. Lüdge, H. Riemann, N. V. Kurlov, D. Borissova, V. Klemm, H. Halloin, P. V. Bm2005. [9] J.L. Chen, G. H. Wu, D. Q. Zhao, S779 - 785, 2001. [1to attain temperatures above 3000 ºC in controlled atmospheres, British Journal of Applied Sciences, 12: 377 – 383,1960.
9
[11] D. S. Robertson, N. Shaw, I. M. Young, Journal of Materials Science, 13: 1986 – 1990, 1978. [12] J. R. Carruthers, G. E. Peterson, and M. Grasso, and P. M. Bridenbaugh, Nonstoichiometry
etani, Growth of FeS single crystals from onstoichiometric melts by normal freezing, Journal of Crystal Growth, 20: 89 – 92, 1973.
e oSb3 grown by vertical Bridgman method, Journal of alloys and Compounds, 386: 228 – 233,
eguez, and J. J. Derby, A Heat shield to control thermal gradients, melt
6] D. E. Eakins, J. B. LeBret, M. G. Norton, D. F. Bahr, and J. Q. Dumm, Electron microscopy
7] P. C. Nordine, J. K. Weber, and J. G. Abadie, Properties of high-temperature melts using
shi, S. Maeda, and K. Nakamura, Prediction of solid-liquid shape during CZ Si rystal growth using experimental and global simulation, Journal of Crystal Growth, 266: 28 -
interface shape in Czochralski oxide growth, ournal of Crystal Growth, 266: 96 - 102, 2004.
. Hozawa, N. Imaishi, K. Shimamura, nd N. Ichinose, Effect of internal radiative heat transfer on spoke pattern on oxide melt surface
o, Numerical simulation of combined flow in Czochralski rystal growth, Journal of Crystal Growth, 179: 309 - 319, 1997.
, Prediction of the growth shape in industrial 300 mm CZ Si crystal growth, Journal of rystal Growth, 266: 34 - 39, 2004.
rystal Growth, 139: 197 - 205, 1994.
and crystal growth of Lithium Nibote, 42 No 5: 1846 – 1851, 1970. [13] T. Takahashi, O. Yamada, and, K. Amn [14] M. Akasak, T. Iida, G. Sakuragi, S. Furuyama, M. Noda, M. Ota, H. Suzuki, H. Sato, Y. Takanshi, and S. Sakuragi, Effects of post-annealing on thermoelectric properties of p-typC2005. [15] J. C. Rojo, E. Diconvection, and interface shape during shouldering in Czochralski oxide growth, Journal of Crystal growth, 200: 329 – 334, 1999. [1of compound oxide laser materials, Proceedings of SPIE 4970, 1, 2003. [1levitation, Pure Appl. Chem, 72 No. 11: 2127 - 2136, 2000. [18] Y. Shiraic33, 2004. [19] O.N. Budenkova, V. M. Mamedov, M. G. Vasiliev, V. S. Yoferev, and Y. N. Makarov, Effect of internal radiation on the crystal-melt J [20] C.J. Jing, A. Hayashi, M. Kobayashi, T. Tsukada, Maon Czochralski crystal growth, Journal of Crystal Growth, 259: 367 - 373, 2003. [21] X. Geng, X. B. Wu, and Z. Y. Cuc [22] T. Wetzel, J. Virbulis, A. Muiznieks, W. V. Ammon, E. Tomzig, G. Raming, and M. WeberC [23] K. Kakimoto, M. Watanabe, M. Eguchi, and T. Hibiya, Flow instability of the melt during Czochralski Si crystal growth, Journal of C
10
[24] Y. Shiraishi, S. Maeda, and K. Nakamura, Prediction of liquid-solid interface shape during CZ Si crystal growth using experimental and global simulation, Journal of Crystal Growth, 266:
8 - 33, 2004.
: 70 – 76, 1983.
. Fukuda, ingle crystal, Journal of
rystal Growth, 180: 546 - 550, 1997.
0.39 Nb2O6 BN) crystals using real-time computer control, Journal of Crystal Growth, 113: 337 - 359,
0] Y. Furukawa, K. Kitamura, E. Suzuki, and K. Niwa, Stoichiometric LiTaO3 single crystal
1] K. Riedling, Autonomous liquid encapsulated Czochralski (LEC) growth of single crystal
2] C. W. Lan, H.J. Chen, C. B. Tsai., Zone-melting Czochralski pulling growth of Bi12SiO20
, S. Q. Wang, N. Konishi, H. Inaba, and J. Harada, Growth habits of 3 and 4-inch
4] X. Lin, Y. Chen, J. Liao, Z. Luo, and Y. Huang, Czochralski growth and spectral properties
Z. You, J. Li, and B. Wu, Crystal growth and spectral properties of aB3O6:Nd crystal, Journal of Crystal growth, 266; 292 – 495, 2004.
rnal of Crystal Growth, 265; 154 – 158, 2004.
2 [25] A. D. W. Jones, Spoke lines, Journal of Crystal Growth, 63 [26] S. Wu, G. Wang, J. Xie, X. Wu, and G. Li, Growth of large birefringent YVO4 crystal, Journal of Crystal Growth, 249: 176 - 178, 2003. [27] S. Wu, G. Wang, and J. Xiea, Growth of high quality and large-sized Nd3+:YVO4 single crystal, Journal of Crystal Growth, 266: 496 - 499, 2004. [28] T. Tsukada, K. Kakinoki, M. Hozawa, N. Imaishi, K. Shimamura, and TNumerical and experimental studies on crack formation in LiNbO3 sC [29] J.P. Wilde, L. Hesselink, Diameter stabilization of Czochralski-grown Sr0.61Ba(S1991. [3growth by double crucible Czochralski method using automatic powder supply system, Journal of Crystal Growth, 197: 889 - 895, 1999. [3GaAs by “intelligent” digital control, Journal of Crystal Growth 89: 435 - 446, 1988. [3single crystals, Journal of Crystal Growth, 245: 56 - 62, 2002. [33] S. Udalangasite single crystals, Journal of Crystal Growth, 237-239: 709 - 713, 2002. [3of Er3+:La2(WO4)3 crystals, Journal of Crystal Growth, 266: 487 – 491, 2004. [35] G. Jia, C. Tu,
3+L [36] C. Tu, Y. Wang, Z. You, J. Li, Z. Zhu, and B. Wu, Growth and spectroscopic characteristics of Ca3Gd(BO3)4:Yb3+ laser crystals, Jou [37] D. C. Stockbarger, The Production of large single crystals of lithium fluoride, Review of Scientific Instruments, 7: 133 – 136, 1936.
11
[38] S. E. King, H. B. Dietrich, R. L. Henry, D. S. Katzer, W. J. Moore, G. W. Philips, and R. C.
, 43, No 3: 1376 – 1380, 1996.
7.
– 155, 1998.
l Growth, 266: 16 – 223, 2004.
T. Zawilski, M. C.C. Custodio, R. C. DeMattei, R. S. Feigelson, Control of growth terface shape using vibroconvective stirring applied to vertical Bridgman growth, Journal of
4] E. D. Bourret, J. B. Guitron, and E. E. Haller, Evaluation of the Mellen “EDG” furnace for
7 – 1224, 1988.
Mania, Development and characterization of zone melt growth GaAs for Gamma-Ray Detectors, IEEE Transaction on Nuclear Science [39] S. Rajendran and R. Mellen, Sr., Advances towards “intelligent processing” of electronic materials, Journal of Crystal Growth, 85: 130 – 135, 198 [40] Chang and W. R. Wilcox, Control of interface shape in the vertical Bridgman-Stockbarger technique, Journal of Crystal Growth: 21, Issue 1: 135 – 140, 1974. [41] C. Barat, T. Duffar, and J-P. Garandet, Estimation of the curvature of the solid-liquid interface during Bridgman crystal growth, Journal of Crystal Growth, 194: 149 [42] R. Ma, H. Zhang, D. J. Larson Jr., and K. C. Mandal, Dynamic of melt-crystal interface and thermal stress in rotational Bridgman crystal growth process, Journal of Crysta2 [43] K.inCrystal Growth, 282: 236 – 250, 2005. [4growth of large diameter GaAs single crystals in the horizontal configuration, Journal of Crystal Growth, 85: 124 – 129, 1987. [45] P. Rudolph, F. M. Kiessling, The Horizontal Bridgman method, Crystal and Research Technology, 23: 120 [46] C. R. Barrett, W.D. Nix, A. S. Tetelman, The Principle of Engineering Materials, Englewood Cliffs, New jersey, Prentice Hall, Inc,1973.
8] D. S. Robertson, N. Show, and I. M. Young, A study of crystals in the cadmium oxide-
rov, Phase equilibrium in the ZnO-TeO2 system,
0] A. Nukui, T. Taniguchi, M. Miyata, Journal of Non-Crystalline Solids, 293-295: 255-260,
1] H. Bürger, K. Kneipp, H. Hobert, and W. Vogel, Glass formation properties and structure of
[47] J. G. Bergman, G.D boyd, and A. Ashken, New nonlinear optical materials: Metal oxides with nonbonded electrons, Journal of Applied Physics, 40, No. 7: 2860-2863, 1969. [4tellurium dioxide system, Journal of Materials Science, 13: 1986-1990, 1978. [49] M.R. Marinov, and V. S. KozhouhaComptes rendus de l’Academei bulgare des Sciences, 25, No 3: 329 – 331, 1972. [52001. [5glasses in the TeO2-ZnO system, Journal of Non-Crystalline Solids, 151: 134 - 142, 1992.
12
[52] M. L. Öveçoğlu M. R. Özlap, G. Özen, F. Altin, and V. Kalem, Crystallization behavior of
3] V. Krämer, and G. Brandt, Structure of Cadmium Tellurate (IV), CdTeO3, Act. Cryst., C41:
nt and D. R. Clarke, On the optical band gap of ZnO, Journal of Applied Physics, 3, No 10:5447 – 5451, 1998.
hesized by plasma-enhanced chemical vapor deposition, Journal of Applied hysics, 95 No 6: 3141-3147, 2004.
some TeO2-ZnO Glasses, Key Engineering Materials, 264-268: 1891 - 1894, 2004. [51152-1154, (1985). [54] V. Srika8 [55] X. Liu, X. Wu, H, Cao, and R. P. H. Chang, Growth mechanism and properties of ZnO nanorods syntP [56] D. R. Lide, Handbook of Chemistry and Physics, 86TH Edition, Taylor & Francis Group Boca Ranton, 2005.
8] T. Lukasiewicz, and A. Majchrowski, Czochralski growth of TeO2 single crystals under conditions of forced convection in the melt, Journal of Crystal Growth, 116: 346-368, 1992. [59] J. G Grabmaier, R. D. Plattner, and ssion of constitutional supercooling in Czochralski-grown paratellurite, Journal of Crystal Growth, 20: 82-88,1973. [60] R. Nayak, A. Nayak, V. Gupta, and K. Sreenivas, Optical interactions in ZnO-TeO2 Bi- layer for AO device applications, IEEE Ultrasonic Symposium-1129, 2003. [61 atterer, M. Wachtler, M. Bettinelli, A. Speghini, and D. Ajò, Optical spectroscopy of lanthanide ions in ZnO-TeO2 glasses, Spectrochimica Acta Part A, 57: 2009 – 2017, 2 [62 ptical second harmonic intensity for ZnO-TeO2 glasses, Journal of Materials Research, 11 No. 10: 2651 – 2 [6 tellurite glasses, Physical Review B, 51:14919 – 14922, 1995. [64] A. N. Moiseev, A. V. Chilyasov, V. V. Dorofeev, O. A. Vostrukhin, E. M. Dianov, B. G. Pl or deposition from organo-metallic compounds, Journal of Optoelectronics and Advanced Materials, 7: 1875 – 1879, 2005.
[57] S. Kumaragurubaran, D. Krishnamurthy, C. Subramanian, and P. Ramasamy, Investigation on the growth of Bi2TeO5 and TeO2 crystals, Journal of Crystal Growth, 197: 210-215, 1999. [5
M. Schieber, Suppre
] R. Rolli, K. G
001.
] Y. Shimizugawa, K. Hirao, The relation between glass structure and poling-induced o
655, 1996.
3] S. Suehara, K. Yamamoto, S. Hishita, T. Aizawa, S. Inoue, and A. Nukui, Bonding nature in
otnichenko, and V. V. Koltashev, Production of TeO2-Zno glasses by chemical vap
13
CHAPTER TWO
EXPERIMENTAL SETUP AND MATERIALS
2.1 Introduction
ZnO-TeO2 and CdO-TeO2 crystals were grown after melting the powders in high purity
containers known as crucibles. Most of these crucibles were pure platinum. Heating th
powders to their melting point was done in a radio frequency coil furnace. The atmosphere in
the furnace can be controlled. Bridgman and Czochr
e
alski were the growth techniques used to
grow
th.
shaft that
holds the seed holder so Czochralski growth could be performed. A manual crank with a
handle was attached to the shaft through a system of gears to reset the vertical position of the
shaft as needed. Both vertical and rotational speeds of the two motors were primarily
controlled by two potentiometers, the rotation motor speed ranged from 2 to 22 rpm, while the
pulling speed ranged from 0.5 mm/hr to 20 mm/hr.
The RF coil was used to heat the material with an operating frequency of 2.5 kHz, and a
maximum power of 45 KVA. The coil’s inside diameter (ID) is 5.6˝ and its total height is 4.8˝.
It has two layers of square tubing and is kept from over heating by a continuous water flow
the crystals. Identification and analysis of the grown crystals were conducted using
XRD, scanning electron microprobe, and by investigating some of the optical and electrical
properties of those crystals.
2.2 The Furnace
The main equipment that was used in this research was a radio frequency (RF) coil
furnace. This furnace was designed and manufactured in China to perform CZ crystal grow
Two high precision motors were mounted on top of the furnace to rotate and pull the
14
system. The coil is placed inside a double layer stainless steel cylindrical chamber. A water
jacket flows between the two layers to prevent the furnace walls from heating. The furnace
has a double walled-window through which water was also running, its importance lies in
enabling the operator to observe the progress of the growth process. See Figure 2.1.
Figure 2.1 Photograph of the furnace with the rotating and pulling motors on top.
A few round holes of different diameters were made in the body of the chamber for gas
inlets and outlets, vacuuming, connecting gauges to determine pressure and oxygen
percentage, and for monitoring the temperature of the hot zone inside the chamber. The
15
electric current passing through the coil was measured by a SR634 AC current transformer.
The signal was then fed into a Hioki 3186 digital multitester where it was processed to return
real the
el
e
i. In
through
B
ne
n runs where very high
In some runs, both types of thermocouples were used at the
same type
J et
t
values of the current. The Hioki 3186 digital multitester was also used to measure
voltage across the RF coils by connecting the voltage terminals of the tester directly in parall
with the coil’s ends. Two analog terminals, located on the backside of Hioki 3186 digital
multitester, were used to interface it to a computer through an interfacing medium.
A roughing pump was connected to vacuum the furnace chamber down to about
20mTorr, whenever needed. A set of manual valves connected the furnace chamber through
metal pipes and hoses to the gas tanks. The valves were installed to control the amount of gas
inside the chamber as required. On the other side of the chamber, an outlet venting valve was
connected to outside the building via a hosepipe to exhaust toxic gases or to vent the furnac
whenever necessary. A Convectron 375 measured the pressure inside the chamber. An analog
pressure gauge was used to measure the pressure when it was between 19.3 psi and 45 ps
addition to that, a Series 2000-percent oxygen analyzer was connected to the furnace
a 0.3" hose to monitor the concentration of oxygen present in the chamber. A set of type
thermocouples were connected to monitor the temperature of different regions of the hot zo
inside the furnace. The thermocouples were fed to the chamber via a vacuum miniature
thermocouple connectors-feed through. In addition to type B thermocouples, type K
thermocouples were also installed to monitor the temperatures i
temperatures are not required.
time. The temperature of the inlet and outlet coil cooling water was measured by a
thermocouple. The recycling water through out the entire furnace was monitored and was s
o cause the furnace to shut down if it exceeds a certain value.
16
2
d from National Instruments, were used. These fieldpoints work as an
interfac
signals. The fieldpoints were powered by a PS-3 National Instruments power supply with
13.8 V and 4 A DC output. They were c r by a RS232 data cable
via PC I /O card.
A set of fieldpoin he furnace.
time
perature of the cooling water of the
coil, pressure inside the furnace, oxygen percentage concentration, rotating and pulling speeds,
Excel spread sheet periodically every certain of most conducted runs, the
prog ord ta ev sec ee .2.
.3 LabView Programming
At the early stages of the research, a LabView program was developed to make the
process of monitoring and controlling the furnace variables easier. A set of fieldpoints (see
table 2.1) purchase
e between the computer and the furnace using small analog voltage and current
onnected to a local compute
Table 2.1 ts used as an interface between the computer and t
Fieldpoint function
1 FP-PS-4 Power supply for the fieldpoints
2 FP-1000 Conn or ects the rest of the fieldpoints with the computer via RS232 connect
3 FP-AI-110 Receives the small analog voltage signals coming from the sensor
4 FP-TB-10 Sends small analog voltage signals to the controllers
5 FP-TC-120 Receives the analog voltage signals that come from the thermocouples
The program currently controls input power, rotating and pulling speeds, it returns real
data of temperatures of the crucible and the hot zone, tem
6 FP-RLY-420 To control the translation and rotation speeds of the shaft
7 FP-Quad-510 Receives the signals from the rotation and the translation motors
current, voltage, apparent coupling power, and mass of the growing crystal. It records data on an
period time. For
ram rec ed da ery 10 onds. S table 2
17
Tabl a fro furn was recorded sec nly rm are The table wa halves paper at.
Pressure Temp.1 Temp.2 Coil Water Te )
e 2.2 Dat m the ace as it every 10 onds. O two the ocouples shown. s split into two to accommodate form
Time Power (Torr) (ºC) (ºC) mp. (ºC Current Time Power
To ensure a uniform and adequate heating process of the crucible inside the furnace, the
crucible was placed inside a special build up which was constructed from high temperature
insulating materials. These insulations were mostly made out of ZrO2, Y2O3, and SiO2/HfO2. A
felt type ZYF-100 made out of ZrO2, Y2O3, and HfO2 was inserted to make the build up uniform
to improve the insula
,
tion construction, and sometimes it was added to make an appropriate
ermal gradient. ZYF-100 type felt properties are listed in table 2.3. The thermal conductivity of
YF-100 type felt insulating material as a function of temperature for different atmospheres is
iven in Figure 2.2.
th
Z
g
18
Table 2.3 Properties of all types of felt insulations. Source: Zircar Zirconia.
YF-10 conductivity as a function of tem ture.
Sour Zircon
Figure 2.2 Felt insulation type Z 0 thermal pera
ce: Zircar ia.
19
The d above floor b cylinder. An alumina
disk of 0.25" thick was placed on top of the cylinder. Liftin was necessary to level
it with th el which is 4.6" above t furnace floor. isk formed
e base for the rest of the build up in all CZ and some Bridgman runs. Properties of alumina are
Tab the hot zone chamber. Source: CoorsTek.
entire build up was raise the furnace y an alumina
g the build up
e RF coil lev he The cylinder and the d
th
given in table 2.4.
le 2.4 Alumina (AD-998) properties. This high purity alumina was used to make the base for the set up in
Property Units Test Value Density gm/cc ASTM-C20 3.92 Crystal Size Microns Thin-Section 6 Water Absorption % ASTM-373 0 Gas Permeability 0 Flexural Strength (MOR), 20 degrees C -- -- 375 (54)
Elastic Modulus, 20 degrees C GPa (psi x 106) ASTM-F417 370 (54) Poisson's Ratio, 20 degrees C -- ASTM-C848 0.22 Compressive Strength MPa(psi x 103) ASTM-C773 2500 (363) Hardness GPa(kg/mm2) KNOOP 1000 gm 14.1 (1440) Rockwell 45 N 83 Tensile Strength, 25 degrees C MPa (psi x 103) ACMA TEST #4 248 (36) Fracture Toughness K(Ic) Mpa m1/2 NOTCHED BEAM 4-5 Thermal Conductivity, 20 degrees C Wm degrees K ASTM-C408 30.0
Coefficient of Thermal -6
Expansion, 25-1000 degrees C 1 x 10 /degrees C ASTM-C372 8.2
Specific Heat, 100 degrees C J/kg*K ASTM-E1269 880 Thermal Shock Resistance, (delta)Tc degrees C NOTE 3 200
Maximum Use Temperature degrees C NO-LOAD COND. 1750 Dielectric Strength ac-kV/mm (acV/mil) ASTM-D116 8.7 (220) Dielectric Constant, 1MHz 25 degrees C ASTM-D150 9.8 Dielectric Loss (tan delta) 1MHz 25 degrees C ASTM-D2520 0.0001
25 degrees C 500 degrees C Volume
Resistivity ohm-cm ohm-cm 1000 degrees C
ohm-cm ASTM-D1829 ASTM-D1829 ASTM-D1829
>1014
2 x 1010
2 x 107
Impingement -- Note 4 0.47 Rubbing -- -- Note 4
2.4.1 Build Up for Czochralski Growth
A large cylinder made by Zircoa Inc. from 9 type insulating material sits on top of the
alumina disk. The cylind its height is 4.2" with an inside diameter
of 4.0”. This insulating c f zirconia and stabilized by 3.5 wt.% calcia, and is
S-362
er’s wall thickness is 0.5" and
ylinder is made out o
20
typically used for high te stal gr rnaces. This material has
good thermal shock prop n. It can survive repeated heating
from room te felt.
Properties of S-3629 cylinders are given in tables 2.5A and 2.5B. To protect the alumina base-
disk from thermal shocks, several insulating materials were placed on top of it to act as a heat
shield. Some of these insulati aterials are recycled powder type ZYFB-3 (see tables 2.6A and
2.6B), and insulating beads type Zirbeads XR which is mostly out of ZrO2. This layer of
insulation could reach as high as 3". On top of the insulation, a then a thin layer of felt was
placed to act as a blank
modified to fit into the setup with a hole in the center.
Table 2.5A S-3629 type insulating cylinders. Source: Zircoa Inc.
Composition 1651
mperature induction heated cry owing fu
erties and great resistance to erosio
mperature to 2000 ˚C or more, especially when used with type ZYF-100
ng m
made
et. On top of all this, comes a 0.5" insulating disk type FBD that was
Stabilizer CaO
Bulk Density (g/cm3) 4.2 Porosity (%) 25 Modulus of Rupture (psi) 2,400 Coefficient of Thermal Expansion RT-600°C (in/in/°C) 8.4 x 10-6
Coefficient of Thermal Expansion RT-1000°C (in/in/°C) 8.0 x 10-6
Coefficient of Thermal Expansion RT-1300°C (in/in/°C) 7.3 x 10-6
Thermal Conductivity (W/m-°K) 800°C 1.2
Table 2.5B Composition of S-3629 type i ulating cylinders. Source: Zircoa Inc. ns
[27] M. Onoe, H. F. Tiersten, and A. H. Meitzler, Shift in the location of resonance frequencies
caused by large electromechanical coupling in thicknesmode resonators, The Journal of the
Acoustical Society of America 35, number 1: 36-42, 1963.
[28] http://www.mt-berlin.com/frames_cryst/descriptions/teo2_pmo.htm accessed on August 16
2006.
aucoma with scanning laser polarimetry, Archives of Ophthalmology, 116 (12):1583-1589,
54
CHAPTER FOUR
same
rials
e
Several mole percentages have been tested for growths, such as 33.3:66.7, and 36.5:63.5. The
choices were limited by several factors, such as the melting temperature, and phase formation.
Microprobe analysis of the grown material of 33.3:66.7 showed it has excess TeO2. The second
mole percentage (36.5:63.5) formed 5-15 mm3 crystals. Optical and electrical measurements
were hard to perform on these crystals. Other growths showed that the 35.5:64.5 mole percentage
gave the best results such that crystals had the least amount of defects. This mole percentage has
a moderate melting point of 650 ˚C, and the resultant material has relatively good-sized crystals
in the range between 50-200 mm3.
an indication of temperature and thermal gradient in the hot zone. These patterns are used by
c
ZnO-TeO2 and CdO-TeO2 GROWTHS
4.1 Zn2Te3O8 Single Crystal Growth
As discussed previously, to pull a single crystal of a certain material, a seed that has the
crystal structure of the projected grown material, has to be used to obtain best results.
Unfortunately, single crystals of Zn2Te3O8 to use as a seed were not found. Many other mate
were examined to be used as a seed, such as tellurium oxide (TeO2) single crystal oriented in th
c-direction, YAG, alumina, zirconia, and a special seed made by dipping the Pt/Rh wire in a
40:60 melt. Sometimes, a platinum wire was used to pull a crystal. Unlike CdO-TeO2 melt, ZnO-
TeO2 melt attaches very well to all these seeds, but the growth of one ingot of a single crystal
was not observed.
As mentioned in Chapter 1, spoke line patterns are used in Nd:YAG melts and other melts as
rystal growers as a starting point of the growth. In ZnO- TeO2 melts, spoke lines were hard to
55
see unless a sudden change in temperature took place. Figure 4.1 shows an example of the spoke
pattern for the ZnO-TeO2 system. This spoke pattern lasted for only 2-3 minutes and disappeared
after the temperature settled down.
ke pattern for 40:60 melt. Similar pattern was observed for other moleFigure 4.1 Spo percentages.
ns were conducted in which 35.5:64.5 melts were pulled into multicrystalline
mat
3
Several ru
erial. A brief description of the procedures and results of these runs are discussed in the
following sections.
4.1.1 (ZTO8)1 Run
In this run, the powders were mixed using a jar mill for 20 hours, and then the mixture was
melted in the 60 cm platinum dish. A platinum foil with a 0.6" hole at the center was used as a
lid to cover the crucible. See Figure 4.2.
Figure 4.2 Top view of the setup used to pull Zn2Te3O8 crystals.
56
A seed of Zn2Te3O8 with minor TeO2 inclusions, obtained from a previous run, was used. A
shown in Fig
s
ure 4.3, the seed was attached to the alumina seed holder by making a notch in both
the seed and the holder and then passing a platinum wire through the two notches. After the
aterial melted, the seed /hr with a
rotation speed of 10 rpm. After a pulling time of 8 hours, the final product was made of
multicrystals that are held together by a powder-like phase. See Figure 4.4. Some single crystals
were extracted and analyzed by a scanning electron microprobe and they were only formed of
Zn2Te3O8. These crystals were too small (5-15 mm3) to carry out further analysis.
m was dipped onto the melt surface, and was pulled at 0.9 mm
ained from a previous growth, attached to the seed holder by platinum wire passing Figure 4.3 Seed obt
through notches made in the seed and the holder.
Figure 4.4 Multicrystalline material resulted from pulling 35.5:64.5 melt for 8 hours. Single crystals were
extracted and scanning electron microprobe shows that they are Zn2Te3O8 single crystals.
57
4.1.2 (ZTO8)2 Run
The same setup and conditions were used as in (ZTO8)1, but the rotation speed was changed
from 10 rpm to 15 rpm. The seed, which was obtained from (ZTO8)1, was attached to the ho
with high temperature cement. After the growth, less powder-like phase was present and a bi
sized mother-crystal was obtained as shown in Figure 4.5. It was noticed that b
lder
gger
igger single
crystals (10-35 mm3) within the mother crystal have formed. Some of the small transparent
multicrystals were isolated, as shown in Figure 4.6. In a whole, the mother-crystal had less
yellow color than (ZTO8)1, this leads to the belief that the yellow color came from the powder
like phase which was not identified. Although a single crystal of Zn2Te3O8 was used as a seed,
microprobe analyses of the small green colored crystals show that they are Zn2Te3O8 with minor
TeO2 inclusions, and this could indicate that the green color comes from these TeO2 inclusions.
Figure 4.7 shows a backscattered electron (BSE) image of the crystals.
Figure 4.5 The mother-crystal obtained in (ZTO8)2, the rotation was increased to 15 rpm rather than 1
0 rpm.
58
Figure 4.6 Some of the single crystals were extracted from the mother crystal shown in Figure 4.5. Scanning
electron microprobe indicates that these single crystals are Zn2Te3O8 with TeO2 inclusions.
Figure 4.7 The crystals are almost wholly Zn Te O with a small proportion of TeO in places. The oran
brown phase is Zn2 3 8, 2 ge-
are
4.1.3 (ZTO )
Another try was done using the 35 mil small platinum crucible, but this time a pure platinum
wire of 0.012″ in diameter was used as a seed. In this run, the inside diameter of the insulating
cylinder was 2.5″. This was a necessary accommodation for the smaller size crucible. See
Figures 4.8A and 4.8B. This time, the powder was pressed in pellets, calcined at 465 ˚C for 24
hours. The pellets were grinded and mixed again in the mill, and after that the resultant mixture
2Te3O8, and the yellow phase is TeO2. The proportions of the two phases in this image not representative, as TeO2 is a small percentage of the entire sample.
8 3
59
was pressed into new pellets, and calcined at 560
as followed to enhance the solid-state reaction of the two powders and to form a stable phase.
ft
ussed in
the runs above, such that small crystals were only obtained with the same abundance of the
powder-like phase. It seems that the above procedure of pressing, mixing, and calcining the
m wa pful in ining ls. ther ses w rrie
9.
˚C for another 24 hours. This procedure [1,2]
w
A er the material has melted, the seed was dipped and pulled at 0.8 mm/hr with a rotation speed
of 20 rpm. The resultant crystal looked almost similar to the previously grown ones disc
ixture s not hel obta better crysta No fur analy ere ca d out. See
Figure 4.
Figure 4.8A Side view for setup used for the small platinum dish.
Figure 4.8B Setup used for the small platinum dish, top view.
60
ire.
4.1.4 (ZTO8)4 Run
In this run, 35.5:64.5 were mixed in the jar mill, and then the material was melted and frozen
in a platinum/gold crucible twice. Melting and freezing more than once helps the material
become more homogeneous and causes the powders to react better with each other [1,2]. The
material was melted and kept at a 700 ºC for one hour. An alumina rod was used as a seed holder
with a seed attached to it. The seed was made out of multicrystals of the same mole percentage;
it was prepared beforehand in a previous run. The seed was dipped onto the melt to start the
growth. The rotation was chosen to be 12 rpm, and the pulling speed was 1.1 mm/hr. The crystal
was pulled for 3-4 hours successfully, but it was noticed that it was not spreading enough, such
that the diameter of the crystal grew less than half the diameter of the crucible. To increase the
diameter of the growing crystal, the temperature was lowered 12 ºC. The crystal started to spread
b
Figure 4.9 Crystal pulled using a platinum w
etter, and the diameter started to increase in a uniform fashion. At this time, the melt was
61
transparent, such that the bottom of the crucible was apparent. No spoke line activities were
noticed. A detailed temperature profile of this run can be seen in Figure 4.10.
Figure 4.10 Temperature of the bottom of the crucible as a function of time.
In a temperature-controlled process, the temperature should be kept at a constant value by
a
power as a function of time. As the crystal grows and some of the melt transforms into
crystalline m
owly. Certain commercial growers, where big silicon or Nd:YAG boules are grown, use the
“decrease in power” to monitor the progress of the growth process. In this run, the decrease in
coupling power was calculated for a 50 gm crystal to be 27.7 Watts. See Figure 4.11B.
djusting the coupling power of the RF coil with the crucible. Figure 4.11A shows the coupling
aterial, the power required to keep the melt at constant temperature decreases
sl
62
Figure 4.11A Coupling power of the RF coil with the crucible as the run develops in time. The large vertical
sudden changes are due to resetting the temperature to a different value.
Figure 4.11B Coupling power of the first
drops slowly as the growth progressesection (shown in Figure 4.11A) of the growth process. The power s. The total power drop for both growth sections is 27.7 watts.
The resultant crystal was a conglomeration of 40-50 small single crystals of different sizes,
ach one is grown in a different direction. The total length of the crystal was 28 mm, and its e
63
diam
On the other hand, a
light decrease in temperature (5-15 ºC) as the growth progresses is a better way to overcome
is problem. This run produced single crystals with sizes ranging between 50-200 mm3.
eter at the bottom is also 28 mm. See Figure 4.12. The crystal was hollow at the bottom
because the heat radiated from the melt, which mostly resulted from the crystallization process
itself, was reflected back to it by the crystal bottom surface which acted as a heat reflector. This
raised the temperature of the melt and caused its meniscus to be convex. A steeper thermal
gradient might be a good way to avoid this from happening in the future.
s
th
Figure 4.12 The 35.5:64.5 as grown crystals.
A comparison of the runs that were conducted to grow Zn2Te3O8 single crystals is shown in
table 4.1 below. Further discussion and analysis of (ZTO8)4 run is made after the table.
64
Table 4.1 A summary of the most important runs used in an attempt to grow Zn2Te3O8 single crystals.
Run # Crucible Seed Pulling speed
(mm/hr) ± 0.1 mm/hr
Rotation speed
± 1 rpm
General Single crystal
size
Microprobe analysis of
single crys
.1.4.1 Discussion and Analysis
I. Single Crystals Analysis
Some of the (ZTO ) run single crystals were isolated. A few of these colorless single crystals
were grinded for powder x-ray diffraction. The resultant x-ray pattern accompanied with a
simulation made by CrystalDiffract 1.3 is shown in Figure 4.13.The pattern was divided into two
segments to obtain a better resolution of the peaks. The resultant data was matched with the PDF
file # 44-0241, known as zinc tellurium or Zn2Te3O8. No other phases were found in this pattern,
but there were two peaks that appear in the simulations but do not appear in the pattern, these
two peaks are marked by arrows. There are also two unidentified peaks that appear in the pattern
at 2 theta = 26 ˚ and 2 theta = 28 ˚. Two more tests were conducted to see if the formed material
is a single crystal or not. The first test is shown in Figure 4.14 where an unpolished single crystal
(rpm) color tals
(ZTO8)1 Zn2Te3O8 with minor TeO2 inclusions
0.9 10 yellow ~5-15 mm3
Zn2Te3O8 only
60 ml Pt dish
(ZTO8)2
60 m Pt dish
Zn2Te3O8 from
light yellow.
crystals are
greenish
mm
Zn2Te3O8 with minor TeO2 inclusions
l (ZTO8)1 0.9 15 Single 10-35 3
(ZTO8)3
A 15 ml
Pt dish platinum wire 0.8 20 yellow and green
~5-20 mm3 N/
(ZTO )
125 ml8 4
wall
from a
run
50-200 2 3 8
y
95%Pt/ 5%Au straight
multicrystals
previous 35.5 1.1 12 clear
white and green mm3
Zn Te OOnly
Done by x-ra
4
8 4
was placed on a polarizer. It shows that the crystal blocks the light at a certain orientation, which
65
means that it is a single crystal. The colors that appear in the photograph might be due to
distortion of light caused by the roughness of crystal surface. The third test was conducted using
a single crystal diffractometer. Figure 4.15 shows one of the peaks of the x-ray pattern which
represents a single crystal peak. The small irregularities that appear on the peak might be due to
defects in the sample, these defects are mainly dislocations. This measurement was also used to
obtain the crystal structure and the lattice constants of the crystal.
Figure 4.13A X-ray diffraction for Zn2Te3O8 a single crystal. The x-ray diffraction was performed using
Siemens D-500 with the following control variables P.V. = 35 kV, I = 30 mA, and CuKα radiation. Data was simulated using CrystalDiffract 1.3 software.
66
Figure 4.13B X . using
Siemens D-500 with the following contr V. = 35 tion. Data was ate alDiffr .3 software
-ray diffraction for Zn Te O a single crystal The x-r2 3 8 ol variables P.d using Cryst
ay diffraction was performed V, I = 30 mA, and CuKα radk
act 1ia
simul .
Figure ph gra singl obtain er pol t. T s of ght be
d storti ght caus rystal ugh4.14 A oto ph of the e crystal ed und arized ligh he region color mi
ue to di on of li ed by c surface ro ness.
67
Figure 4.1 ingl l diffr ter mea t sho 2Te ngl
Som ysta f t 8)4 run were tested for Glow M tro
(GDMS); it shows that these singl tals included some s, s n p 0 Al,
30 Zr, , an Au nexp large p of a in l, came
from the alumin fo e the r was ed fr ind ia a ing
the two powders. Zirconium impurities came fro e insu ter e p
gold c om cru .
II. Electrical a Op ope
Dielec ns t m men
Th ct on the irectio unctio mperature is shown in Figure
4.17. T s f r w d to al to 0 The perature was not noticed in
the tem ture ge n (-7 0) ºC as no sudden change in dielectric constant value
took place. The Curie temperature ot dete to T
5 S e crysta actome suremen ws that Zn 3O8 is a si e crystal.
e cr ls o he (ZTO D geischar ass Spec scopy
e crys impuritie uch as (i pm): 35
72 Pt d 8 . The u ected resence luminum the crysta perhaps,
um il wher powde separat om the gr ing med fter mix
m th lation ma ials, whil latinum and
ame fr the cible
nd tical Pr rties
tric co tan easure t
e diele ric c stant in (001) d n as a f n of te
he los acto as foun be equ .0064. Curie tem
pera ran betwee 5 – 18
was n cted up = 400 ˚C.
68
Figure 4.17 Dielectric constant in the (001) direction as a function of temperature for Zn2Te3O8 single crystal
I-V tests
.
A single crystal sample was tested for a current-voltage measurement in which the resistivity
o
amp
f the crystal was determined. Figure 4.18 shows the relation between the current (I) in Pico
ere and the voltage (V) in volts.
Figure 4.18 Current-voltage relation for Zn2Te3O8 single crystal.
The resistivity, ρ, is calculated to be 1.16 x 10
is found to be 1.4 x 10-13 A. At a voltage difference of -40 V, the current goes to zero and at zero
15 Ω.cm, and the leakage current at V = +80 V
69
voltage, the current measures a non-zero value. This transient state means that the crystal stores
charge when exposed to a voltage difference and keeps this charge when the voltage goes to
ero. Such a performance is more likely to be a capacitor-like behavior.
Transmission and absorption measurement
A single crystal was tested for transmission and absorption. Figure 4.19 shows that the
crystal absorbs the UV radiation up to a cut-off wavelength of 295 nm. The optical band gap was
calculated to be 4.2 ± 0.08 eV. The crystal was solarized for 1 hour by a Xenon vapor lamp to
create internal defects. The solarized [3] crystal spectrum shows a minor difference from the
unsolarized spectrum.
z
Figure 4.19 Absorption spectrum for both Zn2Te3O8 single crystal before and after solarization.
The transmission spectrum for the solarized crystal is shown in Figure 4.20. It does not
illustrate a sharp increase in the transmission at the cut-off wavelength. This might be attributed
to some type of defects in the crystal m ber of Te
om +4 to +2 in response to a missing oxygen atom, which was knocked away by UV radiation
during solarization. Thi ubscript “A” stands
ainly caused by the change in oxidation num
fr
s created defect is identified by [Te2+]A, where the s
70
for near oxygen vacancy. This type of defect is common when a material is exposed to high
energy radiation.
Figure 4.20 Transmission spectrum for Zn2Te3O8 solarized single crystal.
Piezoelectric measurements
A 0.37 mm thick Zn2Te3 if it possesses any
piezoelectric properties. The sample was poled up to 300 V but there was no poling noticed. The
ple was tested for piezoelec
oth
this crystal was verified to be higher than 1.8 by matching oil of the
appropriate refractive index method. This made measuring the birefringence value difficult,
using the available instrumentations at both Washington State University and University of
Idaho. Figure 5.21 demonstrates this phenomenon for this crystal.
O8 single crystal sample was tested to see
sam tricity, but no piezoelectric phenomenon was detected. On the
er hand, poling did not seem to have any effect on the dielectric constant value.
Birefringence
The refractive index for
71
3O8 crystal. Left photograph shows birefringence in the vertical direction, 90 degrees, birefringence took place in the horizontal direction.
IV.
ut β
cted
Figure 4.21 Birefringence of Zn2Te
but when the crystal was rotated
X-ray Parameters
The crystal structure of Zn2Te3O8 was found to be monoclinic, with a, b, and c being (in Å)
12.676, 5.1980, and 11.7810, respectively. In monoclinic crystal structures, α = γ = 90.00 ˚, b
is different from 90.00 ˚. In this present structure, β was found to be 99.60 ˚. This structure
belongs to the space group known as C2/c. A 3D diagram of the crystal structure was constru
using the CrystalMaker 1.3 for Windows. The diagram is shown in Figure 4.22.
72
A summary of some important unit cell and structure parameters are tabulated in the tables belo
Figure 4.22 Zn2Te3O8 crystal structure built using CrystalMaker 1.3 for Windows.
More information about this crystal can be found in appendix 2.
4.2 ZnTeO Single Crystal Growth
Pulling a single crystal of 40:60 ZnO: TeO mole percentage, resulted in a ceramic like
material that breaks easily. The pulled ceramic-like phases were white and of irregular shape, the
material tends to detach from the melt in just 2-3 hours after the growth starts. Dipping the
material once again to resume growth was not successful. Furthermore, the material melts at 715
ºC, which is a high melting temperature at which TeO2 is volatile. The materials composition
3
2
75
will keep changing as the run progresses, and this will lead to a growth of a poor quality crystal
of composition gradient. This argument is apart from the fact that the material has already more
than one phase. See Figure 4.23.
Figure 4.23 40:60 pulled material. The formation of more than one phase and the tendency of the material to
detach from the melt were just a few problems resulting from pulling the material.
About one forth of the runs conducted on the ZnO:TeO2 system were of the mole percentage
40:60. As seen from the phase diagram in chapter 5, 40:60 has more than one phase and goes
through a peritectic transformation. Growth of this material in one single crystal will be difficult.
However, Thermal Gradient Technique was tested to grow a single crystal out of this mole
percentage. Many runs were conducted having the bottom of the crucible cooler than the top
(bottom cooling). It is known that ZnO- TeO2 system is an anomalous material [5,6]. The solid
phase of this system is less dense than the liquid phase, causing various complications during the
growth process. Taking into account the other forces that set the melt into motion, bottom
gradient led to further m
as mixed useless phases. Conversely, when the thermal
op part was cooler (top cooling), much better results were
obt
ixing between different phases. All these runs returned, at their best
results, very small crystals and the rest w
gradient was reversed, such that the t
ained.
76
4.2.2 ZT I
The setup used for this run is shown in Figure 4.24. The main setup elements include a 6
dish-like crucible; the top of it is 0.5″ below the RF level. It was covered with a platinum foil
with a 0.5″ hole in the middle. No insulation was placed on top of the foil; this allowed the to
0 ml
p to
be colder than the bottom. The actual thermal gradient was not measured, but this technique was
ixed in a jar mill for 24 hours and pressed in the hydrostatic press up to 20000
psi, then calcined at 650
used in previous runs and it was noticed that the top starts to crystallize first. (See Figure 5.14).
The powder was m
˚ for 24 hours. The material was then melted at T = 800 ˚C and then the
temperature was lowered to 770 ˚C at which the growth was started by decreasing the
temperature at 1 ºC/hr down to 656 ˚C. The cooling rate was then increased to 10 ºC/hr for 5
hours then increased to 20 ºC/hr down to room temperature. See Figure 4.25.
Figure 4.24 Setup used in Z I run. T
77
Usually, it is difficult to extract the ingot out of the crucible. It was noticed that when the
cooling rate is very slow (1-2 ºC/hr), the ingot just releases in one solid piece after one or two
taps on its bottom. Figure 4.26 shows the material as if it was covered with a powder like solid
Figure 4.25 Temperature as a function of time for ZT I.
phase on the top. The bottom part of the grown material, as seen in Figure 4.27, shows some of
the single crystals coming through the outside layer.
Figure 4.26 The resulting ingot from ZT I run.
78
79
Figure 4.27 A photograph of the bottom of ZT I ingot.
Some of these single crystals were isolated and studied. Figures 4.28 show picture
s of two of
light. The photograph on the right shows that the crystal is almost
tran
Figure 4.28 Single crystals extracted from ZT I ingot.
4.2.2.1 Microprobe Analysis
Some microprobe analyses of the crystals are shown below in table 4.7 and table 4.8; 95% of
the studied crystals are Zn2Te3O8. Figure 4.29 is a representative BSE image of these crystals.
The brown diagonal stripes in the upper left corner of the image are just a grease smear. The
yellow color is Zn2Te3O8, and the small white regions form a pattern and represent TeO2 as a
minor phase with less than 1% zinc presence. Focusing on some of these TeO2 minor phase
these crystals exposed to
sparent to visible light.
regions reveals that a few of them are inhomogeneous as they include a brown phase. This is
shown in Figure 4.30, where the yellow region now is the TeO2 and the brown one is Zn2Te3O8.
Table 4.7 A series of three measurements at different points on the sample. Calculations were made based on the oxide weight percentage and number of oxygen atoms available. At least, 95% of the sample is Zn2Te3O8.
Table 4.8 A series of four measurements at different points on the minor phase. Calculations were made based on the oxide weight percentage and number of oxygen atoms available. The minor phase was found to
be TeO2 with a very few Zn occurrence.
Figure 4.29 A image of Zn2Te3O8 single cryst TeO2 phase.
representative BSE al with some white
80
Figu TeO r phase f m ripe is TeO 3O8.
There is a Zn2Te O8 phase within The TeO ripe. This i omogeneo O type domina he w e cente st th hen ine am
4.2.3
In this run, the powder was prepared in the sam s fo n. fe
include the use of a dish- u cr hic ace low the RF coil level. The
maxi tempe ure of wa and t a er a
hour, and then the temper as de to 7 o l w
developed a steeper therm ient. on t see gle f Z
was dropped on the melt surface to in leat see m . T h
starte decre g the t ture dow ˚C. The crucible was brought to room
temp re at a of 2 ee .31.
re 4.30 2 mino ro Figure 4.30. The yellow stnh
2 aus Te
nd the brown region is Zn2Tent. T3 2
e beam w st 2
is in the “beis not
hit r dot is ju the mach mode”.
ZT II
e way a r ZT I ru A few dif rences
like Pt/A ucible, w h was pl d at 1″ be
mum rat the melt s 820 ˚C was kep t this temp ature for h lf of an
ature w creased 50 ˚C. N id was used in the run hich
al grad In additi o that, a d of a sin crystal o n2Te3O8
itiate nuc ion. The d did not elt or sink he growt
d by asin empera at 1˚/hr n to 640
eratu rate 0 ºC/hr. S Figure 4
81
Figure 4.31 Development of ZT II run as a function of time.
Cooling down was done very slowly (1 ˚C/hr), and it was easy to extract the crystal fr
crucible in one piece. The resulting ingot had two distinct layers, an upper transparent laye
which was 1 cm in thickness and a lower porous brown layer. The two layers were separa
the upper layer was found to form one big single crystal or more with a number of cracks. These
cracks cut throughout the body of the crystal. They mostly formed due to severe thermal stresses
during the cooling process. These single crystals were the source for a number of crack free
smaller single crystals. An example of a polished single crystal is shown in Figure 4.32.
om the
r
ted and
82
Figure 4.32 A single crystal after cutting and polishing resulted from run ZT II.
4.2.3.1 Discussion and Analysis
I. Microprobe and X-ray Analysis
Microprobe analysis of the crystals showed that 99% or more is the single phase of ZnTeO3.
Analysis is shown below in table 4.9, where calculations of the formula were done based on 3
oxygen atoms. Analysis shows that the formula is correct and accurate.
Table 4.9 Microprobe analyses for ZnTeO3 phase.
In some of the crystals, two types of inclusions were found, the first one is the rare phase
ZnTe5O11 and the second one is Zn3TeO6 but with even less abundance.
ith
,
.
s
Powder x-ray diffraction was performed on the single crystals. Data was best matched w
ZnTeO3 PDF card # 44-0240. The pattern has about 40 peaks, more than 34 of them fit exactly
but some small peaks appear in the pattern, and do not appear in the PDF card mentioned above
The identification of some of these peaks was done using CrystalMaker 1.3 simulation, it wa
83
found that some of them belong to Zn2Te3O8, and simulations suggests that these crystals have
less than 3% of this phase, but some peaks were still not identified, they are marked by arrows.
These unidentified peaks might belong to ZnTe
the x-ray
ded into two segments. As x-ray and microprobe analysis showed above,
ZT II produced for the first time ZnTeO3 single crystals. While in ZT I, a combination of
2Te3O8 single crystals and other phases resulted. Although the two powders were prepared in
s melted at a higher temperature of 820 ˚C and was kept there
aterial was melted at 800 ˚C. This 20-degrees difference
in tem
5O11 and Zn3TeO6 phases that were detected by
the microprobe. See appendix 1 for the list of PDF cards that were tested to match with
pattern. Figure 4.33 shows the full pattern of ZnTeO3 matched with its simulation. To better view
the pattern, it was divi
Zn
the same way, the results are different. One explanation for this result is that ZT II run was
exposed to a steeper top thermal gradient when the lid was taken off during the growth process.
In addition to that, the material wa
for a half an hour, while for ZT I, the m
perature might have contributed to the formation and stabilization of the ZnTeO3 phase.
The density of the ZnTeO3 solid phase is the lowest one among ZnO, TeO2 and Zn2Te3O8, if this
is the case, when all these phases are melted, then ZnTeO3 would float. The mass loss for ZT II
was 2.6%, while for ZT I was less than 1%.
84
Figure 4.33A ZnTeO3 powder x-ray pattern and the correspondent simulation.
Figure 4.33B ZnTeO3 powder x-ray pattern and the correspondent simulation.
85
III. Electrical and Optical Properties
Dielectric constant measurement
Dielectric measurement was carried out on the ZnTeO3 single crystal in the (010) direction as
shown in Figure 4.34. The discontinuity in the dielectric value around T = 24 ºC for 1.5 MHz
and 1.0 kHz is due to combining two sets of data collected at two different occasions together. At
room temperature, the average dielectric constant is 14.4 at 1 kHz and the dielectric loss is
0.0375 at 100Hz. A Curie temperature does not seem to exist for the temperature range shown in
Figure 4.35. The sample was also tested up to 500 ˚C, and no Curie temperature was found.
Figure 4.34 Dielectric constant of ZnTeO3 single crystal as a function of temperature at (010).
I-V relation
A current-voltage graph is shown in Figure 4.35. The resistivity, ρ, is calculated to be 3.29 x
014 Ω.cm, and the leakage current at v = +80 V is found to be 8.0 x 10-13A. Comparing these
alues with t 2 3 8 2Te3O8
single crystals are more insulating crystals than ZnTeO3 single crystal.
1
hose obtained for the Zn Te O single crystal, one can conclude that Znv
86
Figure 4.35 I-V relation for ZnTeO3 single crystals.
Absorption measurement
An absorption measurement was performed for the ZnTeO3 single crystal. The absorption
edge is well defined in the UV region as shown in Figure 4.36. The optical band gap was
calculated at λ ~ 300 nm, it is approximately 4.1± 0.08 eV.
Figure 4.36 Absorption spectrum for ZnTeO3 single crystal
87
Piezoelectric measurements
ZnTeO3 single crystals were tested to see if they exhibit piezoelectric properties. A single
crystal was poled at 100 V but there was no poling noticed. Poling voltage was raised in steps to
200 V, but no poling was noticed. Then the poling voltage was raised to 400 V and again, there
was no poling noticed. This is demonstrated in Figure 4.37, where energy (the integration of
leakage current multiplied by applied voltage) was plotted as a function of time before and after
poling. Comparing the two curves, it is noticed that the amount of energy stored in the sample is
very r
ricity as shown in Figure 4.38, but no piezoelectric phenomenon was detected. In
add
small; therefore one can conclude that it did not pole. However, the sample was tested fo
piezoelect
ition to that, poling did not seem to have any effect on the dielectric constant value.
Figure 4.37 Poling did not take place for ZnTeO3 sample since the energy stored in the sample is very small.
88
Figure 4.38 Log impedance as a function of frequency. No piezoelectric effect was noticed.
Birefringence
e higher than 1.8. This made measuring the birefringence value to be difficult using the
ents at both Washington State University and University of Idaho.
IV. X-ray Parameters
The crystal structure of ZnTeO3 was found to be orthorhombic; where a, b, and c are (in Å)
7.327, 6.358, and 12.319, respectively. In orthorhombic crystal structures, α = β = γ = 90.00 º.
This structure belongs to the space group known as Pcab. A 3D diagram of the crystal structure
was constructed using CrystalMaker 1.3 for Windows. The diagram is shown in Figure 4.39.
Using the same method mentioned before, the refractive index for this crystal was verified to
b
available instrum
89
Figure 4.3 iagra s t tru f O s tr s r
n
rtant parameters of the unit cell are tabulated in the tables below:
Table me for 3
p a ( m a ) Ǻ)
9 A d m show he crystal s cture oWi
ZnTedows.
3. It wa cons ucted by Cry talMake 1.3 for
All impo
crystal. 4.10 Unit cell para ters ZnTeO
Al ha (°) Bet °) Gam a (°) (Ǻ) b (Ǻ c (90.00 90.00 90.00 7.360 6.380 12.320
Table 4.11 Some important parameters that have been found using CrystalMaker 1.3 for Windows.
Unit Cell Volume 578.508 Ǻ3
Estimated Density 5533.480 kg/m3
Space Group Pcab Lattice Type P
90
Tab s.
+x +y +z -x -y -z
le 4.12 General equivalent positions. Found by CrystalMaker 1.3 for Window
Table 4.13 Summa of input positional para ta ed from University of Idaho X-ray library.
Fracti al Coordi
ry meter da obtain
on nates Label Site x y z
Occupancy
O (1) O 1 80 .19990 0.22460.472 0 0 O (2) O 1 0.03900 0.3429 0.06710 O (3) O 1 20 .47390 0.40930.156 0 0 Te (1) Te 1 0.06270 0.09260 0.14350 Zn (1) Zn 1 0.10970 0.12180 0.40950
Tab nit cell. Found using CrystalMaker ndows
al C s l Coordi
le 4.14 Listing of atomic coordinates for the first unit cell. Total of 48 atoms exist in the complete u 1.3 for Wi
Fraction oordinate Orthogona nates Label Elmt x y r yor z xo zor O1 O 0.4728 0.1999 0.2246 .521 5433 -0.51018 3 92 2.9O1 O 0.5272 0.8001 0.77 9845 2.11548 -2.17268 54 11.0O1 O 0.0272 0.8001 0.7246 9.94971 -1.4362 -2.17497 O1 O 0 .199 0.27 7067 6.506.9728 0 9 54 4.6 01 -0.50789 O1 O 0.9728 0.3001 0.775 .65439 5.98192 -4.14684 4 9O1 O 0.0272 0.6999 0.2246 4.96598 -.91211 1.46399 O1 0 0.699 0.27 147 .639O .5272 9 54 6.1 3 2 58 1.46628 O1 O 0.4728 0.3001 0.724 .50565 2.43023 -4.14914 6 8O2 O 0.039 0.0671 754 .220.3429 2.0 7 -0 891 1.09172 O2 O 0.961 0.6571 0.9329 5449 .29812. 1 5 72 -3.77457 O2 O 0.461 0.6571 0.5671 .51572 1.99316 -1.18784 8O2 O 0.539 0.3429 0.4329 6.10466 3.07666 -1.49502 O2 O 0.539 0.1571 0.9329 9.91359 2.9334 -6.47721 O2 O 0.461 0.8429 0.0671 4.70678 2.13642 3.79436 O2 O 0.961 0.8429 0.4329 8.73598 5.44198 1.20762 O2 O 0.039 0.1571 0.5671 5.8844 -0.37217 -3.89048 O3 O 0.1562 0.4739 0.4093 5.90313 0.1711 -1.0072 O3 O 0.8438 0.5261 0.5907 8.71725 4.89871 -1.67565 O3 O 0.3438 0.5261 0.9093 10.9464 1.0584 -4.71408 O3 O 0.6562 0.4739 0.0907 3.67398 4.01141 2.03123 O3 O 0.6562 0.0261 0.5907 6.4067 4.21703 -4.18148 O3 O 0.3438 0.9739 0.4093 8.21368 0.85278 1.49862 O3 O 0.8438 0.9739 0.0907 5.98453 4.69309 4.53706 O3 O 0.1562 0.0261 0.9093 8.63585 0.37672 -7.21991 Te1 Te 0.0627 0.0926 0.1435 1.77837 0.21493 -0.69189
[14] D. S. Robertson, N. Shaw, I. M. Young, Journal of Materials Science, Vol. 13, 1986 – 1
1978.
[15] L. L. Patricio, H. Z. Rodolfo, N. Velasco, G. Tarrach, F. Schlaphof
104
CHAPTER FIVE
THE PHASE DIAGRAM
5.1 Introduction
Differences between the phase diagram [1,2] and the experimental results were found.
Results showed that the melting temperature for 21:79 ZnO:TeO2 is higher than 596 ˚C [1] by at
least 20 degrees. Another difference was found when unexpected phases formed as the material
was cooled down from the melt, the formation of these phases were not mentioned in literature.
Conversely, these differences were a motivation to conduct crystal growths on this system. It was
cooling down process takes place. The biggest cha
compound and, eventually, a single crystal can form.
e the
conjectured that there might be a line component forming at a certain composition, when the
llenge was to find the mole percentage of ZnO
and TeO2 at which a stable stoichiometric
The phase diagram shown in Figure 5.1 has two peritectic phase transformations, wher
material transforms from a solid/melt phase into solid. The first transformation takes place at 700
˚C and at 50-mole percentage of TeO2, while the second one takes place at 644 ˚C at 60-TeO2
mole percentage. The eutectic transformation takes place at 596 ˚C when TeO2 is 79-mole
percentage. Some authors [3] refer to 20% mole TeO2 as the eutectic transformation. Two line
components also appear. One at 50% TeO2 and shows that ZnTeO3 compound forms and the
other one appears at 60% TeO2 where Zn2Te3O8 forms. The formation of each of these two
compounds does not come from a single melting phase, but from a combination of (a) melt
phase/s and a solid phase. As a result, tuning the melt and the solid phase to the exact required
composition to form the line component would be extremely difficult. This explains how in
crystal growth, it was very hard to obtain a single crystal. But luckily, the thermodynamics were,
105
occasionally, correct for some single crystals to grow, as occurred in the growth of single
crystals of ZnTeO3 and Zn2Te3O8. Unfortunately, the existence of these “correct
thermodynamic” conditions would not stay for a long time. The combination of composition,
temperature, and cooling rate should be optimized to continue the growth of a single crystal. As
the single crystal growth continues, the composition of the melt required for a single crystal
growth changes, and this causes the composition to shift either to the left or to the right in the
phase diagram. In any case, another/ other phase/s will appear. Consequently, this will terminate
the growth of the single crystal. Another growth difficulty includes the existence of a solid phase
above each line component. At 450 ˚C, the solid phase that has formed at higher temperatures
will go through another phase transformation and polymorphism will exist, such that more than
horizontally up to 100% TeO . At this line, α-Zn O transforms into β-Zn Te O . The phase
diagram in Figure 5.1 shows all possible stable phases that could form, accompanied with their
required ZnO mole percentages.
one solid phase will coexist. The polymorphous line starts at 50% mole TeO2 and extends
2 3 8 2 3 82Te
106
Figure 5.1 Part of the phase diagram of the ZnO-TeO2 system. The phase diagram was taken from Bürger e
5.2 ZnO–Te
t al after modification. Arrows show where potential line components would form.
O2 Phase Diagram
The study of the phase diagram was mostly conducted by two methods, powder x-ray
diffraction and scanning electron mic ercentages were mixed, calcined at
a temperature below the melting point of the mixture, melted at different temperatures above the
melting point, or pulled. . Some mole percentages went through more than one of the processes
mentioned above. In general, powders were mixed as mentioned before. The most important
mole percentages that were studied are covered in more detail in the following sections.
roprobe. Several mole p
107
5.2.1 ZnO:TeO2 - 9:91
This mole percentage was studied to investigate the type of phases that appear at a
considerably low ZnO mole percentage. 9:91 is about halfway between the eutectic
transformation and pure TeO2 in the phase diagram. The powder was mixed for 19 hours then
placed in the platinum dish. It was heated, using the temperature-controlled process, up to 500 ˚C
at 30 ˚/hr, and then up to 760 ˚C at 50 ˚/hr. The furnace chamber had a continuous flow of
oxygen at atmospheric pressure during the run. The crucible was immediately cooled down to
room temperature at 50 ˚/hr. Figure 5.2 shows the time development of the run for both
temperature and coupling power. A small portion of the cooling down process is shown. Two
distinct dips in the power are noticed; each dip marks the formation of a phase. The first dip is at
T
phase diagram, it showed that the phase diagram is accurate when concerning the existence of
t
phase diagram suggests. The slope (m2) of the middle section is less than the other two slopes,
since the material is giving away heat as the phase is forming.
= 707 ˚C and the second one is at T = 616 ˚C. When these values were compared with the
wo phases, but both phases were found to form at an average of 15 ˚C higher than what the
Figure 5.2 Cooling down the crucible in a temperature-controlled process. Each dip in the power curve marks
the formation of a different phase.
108
As seen in Figure 5.3, the surface of the resultant material seemed homogeneous except for
some limited areas. These areas are distributed in a random pattern, concaved upwards, and are
of a lighter color than the other material. They seemed to belong to a different phase. Two
samples were studied, one was cut from the homogeneous region, and the other one was from the
non-homogeneous region. Scanning electron microprobe analyses were performed on these two
samples.
Figure 5.3 The resultant 9:91 material.
Figure 5.4 is a BSE representative image of the non-homogeneous area where two phases can
be seen. The large rectangular yellow areas are TeO . The orange part is a fine-grained
assemblage of TeO and Zn Te O . The formula is calculated to be Zn Te O , with a total of
99.6%.
The microprobe analysis of the homogeneous region shows the same two phases found
earlier in the non-homogeneous areas, but the way these phases are distributed is different. In
owever, the Zn2Te3O8 in this case gives somewhat low totals, averaging 97%, and the formula
Zn1.90Te3.05O8.
2
2 2 3 8 1.93 3.04 8
Figure 5.5, the bright yellow solid phase is TeO2, and the other part contains TeO2 and Zn2Te3O8.
H
is
109
Figure 5.4 A BSE image of one of the non-homogeneous areas, as seen by scanning electron microprobe
representative image.
Figure 5.5 A BSE representative image of a homogeneous region. The yellow represents TeO2 and the ora
contains both TeOnge
sult
rectangular shape, but it does not appear the same way in the homogeneous region.
2 and Zn2Te3O8.
Although the physical look of the homogeneous region and the non-homogeneous region is
different, it appears that they are formed of the same two phases, TeO2 and Zn2Te3O8. This re
agrees with the phase diagram. The difference in shape and appearance between the two regions
might be attributed to both the percentage of presence and the distribution of each phase in each
region. For example, TeO2 in the non-homogeneous region appears to be of a uniform
110
5.2.2 ZnO:TeO2 - 16.7: 83.3
Ideally, if five moles of TeO2 were mixed with one mole of ZnO and they reacted, it should
give ZnTe5O11, following the reaction
11525 OZnTeTeOZnO →+
The upper chemical equation is balanced and the oxidation numbers for Zn and Te are +2 and
+4, respectively. The question was whether this phase could be formed using the 16.7:83.3 mole
percentage. To answer this question, this mole percentage of ZnO and TeO2 was mixed
thoroughly in the jar mill for 20 hours, and then pressed into pellets and calcined at 515 ˚C for 24
hours. The pellets were crushed and mixed again for another 20 hours, pressed again and
calcined for an extra 20 hours. The resultant material was studied by x-ray and scanning electron
microprobe. Figure 5.6 shows a representative BSE image of the powder using the scanning
microprobe. The elemental analysis confirms the presence of two phases, the bright one is TeO2
and the brown phase is Zn2Te3O8 with sums of 101.163 and 100.797, respectively. The dark
areas are due to the glass substrate.
Figure 5.6 Microprobe image of 16.7:83.3 calcined powder. The black area is the glass substrate.
X-ray analysis is in agreement with the results obtained by the microprobe. The resultant x-
ray pattern of the 16.7:83.3 is shown in Figure 5.7A. This pattern was normalized and compared
111
with the TeO2 and Zn2Te3O8 PDF cards as shown in Figure 5.7B. To better understand the
contribution of each phase to this pattern, data were compared with a simulation generated by
CrystalDiffract 1.3. It shows that 60% of the sample is TeO2 and the rest is Zn2Te3O8, and no
ZnTeO3 e
ea
was found. The pattern was divided into two segments to obtain better resolution of th
ks. The fit confirms that there is only two phases as shown in Figures 5.8A and 5.8B. p
Figure 5.7A X-ray pattern for 16.7:83.3.
112
Figure 5.7B X-ray p 16.7:83.3 compared with PDF patterns of TeO2 and Zn2Te3 m
on o and Zn2 8 is 58.2 1.75, r vely.
attern for co siti
O8. Phase diagrampo f TeO2 Te3O 5 and 4 especti
Figure 5.8A X-ray pattern of 16.7:83.3 and the generated patterns for TeO2 and Zn2Te3O8.
113
Fi re 5.8B X-ray pattern of 16.7:83.3 and the generated patterns for TeO2 and Zn2Te3O8. The fit shows a
perfect match.
5.2.3 ZnO:TeO2 - 21:79
This mole percentage was the most studied composition of the ZnO and TeO2 system. In the
phase diagram, it is the point where a eutectic phase transformation takes place at the lowest
temperature. As shown in the phase diagram at the beginning of this chapter, a straight horizontal
line extends from the line component of Zn2Te3O8 up to the TeO2 pure composition. The
material was calcined, melted, and pulled. Many tests at this composition were conducted. Most
of these tests were done using x-ray diffraction and Microprobe analysis. These tests and their
results will be discussed in further detail in the following sections.
5.2.3.1 21:79 – A general look
After thoroughly mixing the two powders, the mixture was placed in a platinum / gold
straight wall crucible and heated in the RF furnace up to 670 ˚C. At this point, no smoke was
observed and melt was extremely transparent. A TeO2 semi cylindrical seed was dipped onto the
gu
114
m
texture as show
A sample was prepared for the electron microprobe. A representative image of the sample is
hown in Figure 5.10, where the dark phase is Zn Te3O8, and the brighter yellow phase is TeO2
with minor ted
uniformly. Using Clemex Vision Analysis software, Zn2Te3O8 was found to form 52.5% and
TeO
elt surface and pulled at 14 mm/hr. The pulled material was dark grey color and had a ceramic
n in Figure 5.9 below.
s 2
Zn (usually less than 1%). The image shows that the two phases are distribu
2 forms 46.6%, the rest is cracks and Zn.
Figure 5.9 Pulled 21:79 material.
s found to foFigure 5.10 BSE image of led material. Zn2Te3O8 wa rm 52.5% and TeO2 forms
.6%, the rest is cracks and som Work conducted on this ma story of the has a great effect on the
results. This was noticed on almost every mole percentage that was tried. Any process that takes
place on the material can change the results noticeably. To show this, several runs were
conducted on 21:79 same material. The results are summarized in table 5.1. Photographs of each
21:79 pul46 e Zn.
terial shows that the hi material
115
resultant material are also shown. See Figure 5.11. No x-ray or probe results are shown at
this time. All these runs were done with the same powder, which was mixed, calcined, and
melted in a standard b to these runs.
Table 5.1 Summary of some details of 21:79 runs using the same powder.
Run Crucible (Platinum) ere
Max.Temp
(ºC) (º/hour) ass loss% Heating/ cooling control
micro
ox furnace prior
setup atmosph
.
cooling rate Accumulated
m
1 Big dish 665 Temp. No lid
Oxygen 7 0.06
2 Big di 685 12 0.0 Temp. sh No lid Air
3 Small dish Lid 710 30 0.0 Temp. air
4 Small dish
lation on top, hole on bottom, to
ermal 731 2.3 0.29 Power enhance th
Lid and insu
gradient air
5 Small
Lid and insulation on top, hole on bottom, to
gradient air
Power dish enhance thermal 821 23 0.59
6 Small dish
Lid and insulation on
enhance thermal gradient
air
678 53 N/A Power top, hole on bottom, to
Figure 5.11 Difference in color and physical appearance for several 21:79 runs
Run 1 Run 2 Run 3
Run 4 Run 5 Run 6
116
5.2.3.2 Phase formation in 21:79
At the eutectic transformation in the phase diagram, where the melt transforms into two
different solid phases, namely: TeO2 and Zn2Te3O8, the transformation takes place at the same
temperature, and at the same time. Any deviation from the eutectic percentages will cause time
and temperature dispersion in the formation of these two solid phases. This will be a sign if the
melt loses some of its constituents due to TeO2 evaporation. This has been tested in some runs,
such that in a temperature-controlled run, a new calcined powder was melted. During the cooling
down process, one dip in the power was noticed marking the formation of the two phases at the
same time. This is shown in Figure 5.12. The temperature path as a function of time is also
shown.
s, there was a mass loss in TeO2 due to evaporation. The crucible cooling down rate was
set
Figure 5.12 Coupling power and temperature as a function of time.
Two more runs were conducted on the same material resulted from the run above. In these
two run
to be 50 ºC/hour in a power-controlled run. At the cooling down process, two close peaks
appear in the temperature curve. The appearance of these two peaks is a sign that the material is
not on the eutectic transformation any more, but shifted towards a lower TeO2 composition. See
117
Figure 5.13. The Figure is accompanied by the ZnO-TeO2 phase diagram, where the arrow
indicates the possible place at which the cooling process took place. The arrow position is an
exaggeration of the situation; it should be closer to the eutectic reaction.
Figure 5.13 Cooling down the melt of 21:79 after losing some TeO2 due to evaporation. The two humps c
5.2.3.3 ZnO:TeO
an be seen indicating the formation of two phases at two different temperatures.
ontrolled run was used as a trial to melt a calcined 21:79 powder. A
relatively slow cooling rate and temperature gradient were employed in an attempt to see the way
the material starts to crystallize. For this purpose, the material was melted in the crucible with a
platinum lid on top. When the top temperature value reached 625 ˚C, the lid was taken off and a
small Zn2Te3O8 crystal was carefully dropped in the middle of the melt surface. The crystal did
not sink, which indicates that the crystal density is less than the melt density. At this time, the
bottom thermocouple was reading 685 ˚C, with a thermal gradient of 30-degrees/ cm. After a few
minutes, straight, radial, and uniform rays started to form and grew bigger to form a star like
2 - 21:79 Top Cooling
A temperature-c
118
shape. This layer looked thin and partly melted. A few minutes later, another layer started to
form and grow on top of the star- like layer, as if the star-like layer forms the foundation of that
second top layer. The growth process of both layers extended until the star-like layer got close to
the crucible edge, where the temperature is very high for more growth to take place. See Figure
5.14.
in
minutes.
After a few minutes the growth stopped and the solid phase floated on top of the surface, as
/hour. Samples were taken from three different parts of the resultant material. One sample was
ingot, and the third one was taken from the bottom center of the ingot.
Figure 5.14 A series of photographs of 21:79 melt after dropping a Zn2Te3O8 seed on the surface. Time is
shown in Figure 5.15. Afterwards, the crucible was cooled down to room temperature at 25 ºC
taken from the top middle part of the ingot, the second sample was taken from top side of the
t = o t = 1 t = 2
t = 3 t = 4 t = 4.5
119
Figure 5.15 The right photograph shows a close up look at the growing material. A stabled solid material floats on top of the melt.
The purpose behind choosing these three different areas is to see if there was an axial or a
radial gradient in the composition of the forming phases, and, to see if the percentage of the
forming phases were the same across the ingot. Microprobe analysis shows that two phases,
TeO and Zn Te O , exist in the resultant material with roughly the same area percentage in all
the three samples. X-ray diffraction of the samples shows the presence of the two phases. These
results agree with those obtained by the microprobe. Figure 5.16 is an overlap of the three
After 5 minutes After 5 minutes
seed
2 2 3 8
different patterns. These patterns are alike, such that the peaks positions are similar in all three
patterns. However, the s compared to a
simulated one generated by CrystalDiffract as shown in Figures 5.17A, 5.17B, and 5.17C. The
simulation shows that the TeO2 phase is less abundant near the top area of the ingot, but it has
more presence than the Zn2Te3O8 at the bottom. This might be attributed to the fact that the TeO2
phase is denser than the Zn2Te3O8 phase. The percentage of each phase is shown in the Figures.
The patterns were also compared with the PDF cards to see which phases fit the patterns the best.
It was found that TeO2 of PDF card# 42-1365 and Zn2Te3O8 of PDF card# 44-0241 are the best
candidates.
intensities of many peaks vary. Each x-ray pattern wa
120
Figure 5.16 X-ray patterns for 21:79 with top cooling at 25 degrees /hour. All tested parts of the ingot show
almost the same pattern.
Figure 5.17A X-ray pattern of a 21:79 sample taken from the top side area of the ingot. The red pattern is a
simulation of a mixture of TeO2 and Zn2Te3O8.
121
Figure 5.17B X-ray pattern of a 21:79 sample taken from the top middle area of the ingot. The red pattern is
a simulation of a mixture of TeO2 and Zn2Te3O8.
Figure 5.17C X-ray pattern of a 21:79 sample taken from the bottom middle area of the ingot. The red
pattern is a simulation of a mixture of TeO2 and Zn2Te3O8.
122
Another similar run to the one above was also conducted. This time the cooling rate was
much slower (1 ºC/ hour). As found in the previous run, both microprobe and x-ray show that the
resulting material has only TeO2 and Zn2Te3O8. The pattern was also compared with generated
x-ray patterns using CrystalDiffract 1.3. The pattern was divided into two segments to obtain a
better view of the peaks. The fit confirms that there are only two phases, namely 56% TeO2 and
44% Zn2Te3O8 as shown in Figures 5.18A and 5.18B. The sample was taken from the center of
the ingot. This explains why the TeO2 percentage is high in compliance with the previous result.
Figure 5.18A X-ray pattern of 21:79 material and the simulated patterns for the TeO2 and Zn2Te3O8 mixture.
123
Figure 5.18B X-ray pattern of 21:79 and the generated patterns for TeO2 and Zn2Te3O8. The fit shows a
perfect match.
The way the sample was prepared for x-ray diffraction had a considerable effect on the x-ray
da
nsities of some peaks were
ifferent. This means that grinding hard or gently does not have an effect on producing new
hases or terminating existent ones. However, it might have some effect on the orientation of the
crystals in the powder, such that some planes are more or less probable to interact with x-rays.
See Figure 5.19. Furthermore, when the powder was grinded with bare hands, a distinctive
pattern was obtained that does not match any PDF card. Figure 5.20 shows this pattern. This
means that pressure could be a factor in forming a new phase.
ta results. In the run that was just mentioned above, two powder samples were prepared, one
was grinded hard by the pestle and the other was grinded very softly. Although there were no
differences in the peaks positions between the two patterns, the inte
d
p
124
Figure 5.19 21:79 material, where two patterns are shown. One pattern is of a powder that was grinded hard,
and the other one is of a powder that was grinded gently.
fpattern.
was used to melt the
Figure 5.20 21:79 material, one grinded by pestle and mortar, and the other one grinded by the tips of ingers. The peaks appearing in the hand-grinded pattern were not identified. No PDF card has a similar
In all runs of 21:79 mole percentage mentioned above, the temperature was raised above the
melting point by 30 degrees. In the following run, a standard box furnace
125
material, which was placed in a small platinum crucible. The temperature was raised to 600 ˚C at
elt the material without going any higher in temperature in the melt phase.
this would affect the forming phases. The material was then cooled
dow
ooked dark brown. However, Figure
5.21 confirms that the forming two phases are the same two phases that always have formed in
21:79, namely, TeO2 and Zn2Te3O8.
once, and then it was raised by 1 ˚C every 15-25 minutes, until the material melted at around 621
˚C. The idea was to m
This was done to see how
n to 595 ˚C in two hour, then down to 584 ˚C in 20 minutes. After that, the crucible was
quenched in air at room temperature. The resulting material l
Figure 5.21 X-ray patterns for the 21:79 material which was barely melted in a standard box furnace.
CrystalDiffract 1.3 was used to simulate these r percentages in the
samp
Patterns show that the material has two phases, namely TeO2 and Zn2Te3O8.
two phases and obtain thei
le, which were 47% TeO2 and 53% is Zn2Te3O8. No other phases were found. This is
shown in Figures 5.22A and 5.22B.
126
Figure 5.22A X-ray pattern and simulation for 21:79 material melted at 621 ºC.
Figure 5.22B X-ray pattern and simulation for 21:79 material melted at 621 ˚C.
So far, the 21:79 mole percentage material returned only two phases. All powder x-ray
patterns and scannin 8 are the only two
hases that appear in this mole percentage. Melting at different temperatures, pulling, or
alcining led to the formation of the two phases mentioned above.
g electron microprobes confirmed that TeO2 and Zn2Te3O
p
c
127
5.2.3.4 ZnTe6O13 C
When the powder was calcined, then melt se es, t 770
˚C in a power-controlled run, and cooled down at 2 t of 60
˚/cm as in Bridgman t ique, a br new phas ed. If this procedure is followed, see
table O13 fo instead o O2. This i irst repo
Max. Mass
(%)
rystal from 21:79 Mole Percentage
ed and frozen veral tim and then melted a
˚/hr with an axial temperature gradien
echn and e form
5.2, ZnTe6 rms f Te s the f rt of the existence of this phase.
Table 5.2 Processes that led to the formation of the new phase.
process setup temp. (ºC)
cooling rate (K/hour) loss
Type of control
Notes
calcining standard box furnace
500 for 24 hours quench zero N/A quench in air, no
lid
melting n lid R.F. 666 7 0.06 temp.
air, and Oo
furnace Torr
2 flow above 999
melting R.F. 685 12 0.0 temp. air only no lid
furnace
melting
R.F. furnace
650 30 0.0 temp. air with a lid
melting with a lid
R.F. furnace
780, briefly 48 0.01 power
Start cooling at 720 (ºC)
air
A chunk of the resultant material was crushed into pieces. Some of these pieces had a light
brown color and sharp edges with a hard texture. These small pieces were actually single crystals
of the new phase. They were isolated and analyzed using single crystal x-ray diffraction.
The new phase crystallizes in the hexagonal space group R-3. Figure 5.23 shows a fragment
of the unit cell and a calculated powder x-ray diffraction pattern is shown in Figure 5.24. There
are two unique Te atoms in the asymmetric unit. Both are four coordinate trigonal bipyramidal,
i.e. TeO42- with a stereochemically active lone pair of electrons, a common motif in tellurate
structures. The environment around Te1, although compressed, is more similar to that in α-TeO2
(Te-O 1.919 and 2.08 ion around Te2 is 7 Å; apical O-Te-O 163.9º) [4]. However, the coordinat
128
similar to the TeO3+1 coordination of e.g. CuTe2O [5] and CuTe3O8 [6] with three Te-O
istances 1.857(4), 1.922(4), 2.026(4)Å, and the fourth significantly longer at 2.204(4)Å. The
coordination sphere around the Zn atom is a highly distorted octahedron with trans O-Zn-O
angles of ca. 163º. Three of the oxygen atoms are linked to Te1 units and the other three are
linked to Te2 units. The three Te2 units and the Zn1 form an adamantly type substructure. The
three Te1 units and Zn1 form a “paddle wheel” arrangement with oxygen bridged O2-Te1-O2
atoms (see Figure 5.23). The complete packing superstructure is shown in appendix 4 and
consists of a bilayer of tellurium oxide linked by the Te1-O2 bridging units. These bilayers are
connected via the Zn atoms. Table 5.3 shows Summary of input positional parameter data.
Crystal data and refinement parameters can be found in table 5.4. More information about the
crystal structure and bonding can be found in appendix 4.
5
d
129
Figure 5.23 A diagram shows a part of the unit cell of ZnTe6O13. Selected bond lengths and angles: Te1-O1 2.1244(7); Te1-O2 1.936(4); Te1-O3 1.851(4); Te1-O2a 2.168(4); Te2-O3 2.204(4); Te2-O4 1.922(4); Te2-O5
Table 5.4 Crystal data and refinement parameters for the ZnTe6O13 phase.
Formula ZnTe6O13
formula wt 1038.97
crystal system Hexagonal
Space group R -3
a (Å) 10.1283(9)
b (Å) 10.1283(9)
c (Å) 18.948(3)
V (Å3) 1683.3(3)
Z 6
T (K) 86(2)
λ (Å) 0.71073
ρ calc (Mg/m3) 6.149
µ (mm-1) 17.551
F(000) 2676
crystal size (mm3) 0.13 x 0.06 x 0.02
θ range(°) 2.56 to 25.23
Index ranges -12≤h≤12, -12≤k≤12, -22≤l≤22 Refl. Coll 8ected 273
Indep. Refle 9 [R(int) = 0.033ctions 67 2]
da traints/ . 679 / 6 / ta/res param 62
GOF 1.123
*R1 [I>2σ (I)] 0.0200
*wR2 [I>2σ( )] 0.0593 I
Larges . peak, Å-3) 0.726 - t diff hole (e 0.690
1 = ΣFo - ΣFo; wR Σ F 2 - Fc2)2]/ Σ
Powder x-ray diffraction was also carried out for the material. Th
was compared with a simulation generated b
*R Fc / 2 = [ w( o [w(Fo
e resulting x-ray pattern
2)2]1/2
y CrystalDiffract 1.3. The simulation shows that
132
there is no o e o O2 at uch th aterial is
Z O8. T ow Figu 25.
ccurrenc f Te all, s at 54% of the m ZnTe6O13 and 46% is
n2Te3 his is sh n in re 5.
Figure 5.25 X da Te6O13 compared with tion data of the Zn2Te d Zn
u
It lik e Te as re wit ZnO Zn an ormed a ZnTe6 ha
might be attributed to the ma has hea up and led d n ma
2Te3 8 into , or ZnO and Te
especially when the temperature was 770 ºC to achieve that. By that ti , and due to a sl
li n, t free e y of yste shed o for nTe6O stea TeO2
excess presence of oxygen in a previous run might h e also layed a role in this unex
resu
-ray ta for Zn the simula 3O a8 n Te6O13 mixt re.
looks e th O2 h a dcte h the or d f O13 p se. This
fact that the terial been ted coo ow n s, y time
and this might have led to the decomposition of part of Zn O Zn O , 2
me ow
coo ng dow he nerg the s m pu it t m Z 13 in d of . The
av p pected
lt.
133
Scanning electron microprobe was performed. A BSE image is shown in Figure 5.26. At
first, it was thought that 6O phase has formed, but analysis based on 11 atom
return r c lat the Z (0.85 d Te .074) s num rs. The total summation
oms of oxygen, the average number
of Zn atoms was 1.007 and for Te atoms was 5.997. For this suggested number of oxygen atoms,
s was 20.003. Scanning microprobe shows no occurrence for TeO2
phase, and this is in agreement with the x-ray analysis.
a ZnTe 11 s of oxygen
ed poo alcu ions of n 2) an (5 atom be
of atoms was 16.926. When analysis was done based on 13 at
the total summation of atom
Figure 5.26 BSE representative image. The yellow area is ZnTe6O13 and the brown area is Zn2Te3O8. The
black area is just the glass substrate
2
5.2.3.5 Reaction Detection of 21:79
A new mixed powder of 21:79 was put in a platinum dish and heated up in a power-
controlled run. The crucible was covered using a platinum lid with a hole in the center. The hole
was made to enable the thermocouple to reach the powder and its tip to be immersed in it. As the
material was heated up to 505 ˚C, a sudden peak in temperature took place at 446 ˚C. At this
temperature, it is suggested that ZnO and TeO reacted with each other as shown in Figure 5.27.
134
The temperature was kept at 505 ˚C for 24 hours, and then the furnace was shut down. The
temperature dropping rate was monitored, and no sudden change in temperature measurement
was noticed. This means that the peak appeared in the heating section is due to a reaction not due
to a phase change.
Figure 5.27 Reaction detection for 21:79 mixed powder.
Non ambient x-ray diffraction was utilized to detect for the reaction. A new 21:79 powder
holder was placed inside Anton Paar
rns were collected at several temperatures starting from
ing
.
ow that the mixture starts to react between 400 ˚C and 450 ˚C or
mixture was placed inside a small alumina holder. This
Chamber (HTK 1200), and x-ray patte
room temperature and ending at 585 ˚C. X-ray patterns were collected at the follow
temperatures in ºC: 25, 250, 325, 370, 400, 450, 500, 550, and at 585. For simplicity, only
patterns collected at 25 ºC, 400 ºC, 450 ºC, and 585 ºC are shown. See Figures 5.28A and 5.28D
The series of x-ray patterns sh
135
before. As a matter of fact, some ZnO peaks intensities start to decrease at T = 325 ˚C, and some
little peaks for Zn2Te3O8 start to form slowly. Figures 5.28B and 5.28C show the final step for
this process. The x-ray pattern collected at 450 ˚C shows that ZnO powder has reacted
completely forming Zn2Te3O8 replaced it. Although a total reaction of ZnO does not take place
until the temperature is 450 ˚C, the formation of Zn2Te3O8 starts at a lower temperature. This
contradicts the previous result, which shows that the reaction takes place at one singular
temperature (as shown before in Figure 5.27). The non-ambient temperature x-ray pattern shows
that this process takes place over a wide range of temperatures. Figure 5.28A shows three
p
e e
with the
ing that
atterns at room temperature, one for TeO2 PDF, one for the ZnO powder collected earlier, and
th third one is for the 21:79 powder. In Figure 5.28D, The PDF of both TeO2 and Zn2Te3O8 ar
overlapped with the collected pattern at T = 585 ˚C. Notice that there is no overlapping
ZnO pattern, since it had completely reacted. It can also be noticed for the collected pattern at
this high temperature that pattern looks compressed from both ends by about 0.449 degrees. This
is attributed to the high temperature, which affects the distance between the planes of the crystals
and hence affects the angle at which x-ray diffraction takes place. It is worth mention
these patterns were collected using a CoKα x-ray, and then they were transformed into CuKα
patterns by using Bragg’s equation. This was done to compare these patterns with PDF cards.
136
smoothing. The patterns were collected using a cobalt tube. Figure 5.28A X-ray patterns for TeO2 PDF, ZnO powder, and 21:79 mixture at room temperature after
Figure 5.28B X-ray patterns for TeO2 PDF, ZnO Powder, and 21:79 mixture T = 400 ˚C after smoothing. The
patterns were collected using a cobalt tube.
137
Figure 5.28C X-ray patterns for TeO2 PDF, ZnO Powder, and 21:79 mixture T = 450 ˚C after smoothing. The
patterns were collected using a cobalt tube.
Figure 5.28D X-ray patterns for TeO2 PDF, ZnO Powder, and 21:79 mixture at T = 585 ˚C after smoothing.
The patterns were collected using a cobalt tube.
138
5.2
t
s undevitrified up to 320 ˚C. At 410 ˚C, peaks start to appear indicating the
beg
, a
s temperature took place. Most of the
changes can be investigated by only focusing on 2-theta values between 25 and 40 degrees as
shown in Figure 5.29A. This Figure shows that at T = 340 ˚C, peaks of Zn2Te3O8 phase start to
appear and two more unidentified peaks marked with arrows appear alongside. This might mean
that as glass is crystallizing, it transforms into an intermediate Zn2Te3O8 phase mixed with small
proportions of TeO2, as some peaks of this phase appear too. As the temperature increases to T =
360 ˚C, the pattern does not change, but it is noticed that the TeO2 peaks become more defined.
The few peaks that do not match the PDF patterns might belong to intermediate phases of both
TeO2 and Zn2Te3O8, but mostly of Zn2Te3O8. As temperature increases, the two phases start to
leave their intermediate appearance to form their most regular phases. This behavior continues
p to T = 415 ˚C.
.3.6 Devitrification of 21:79 glass
A similar test of the non-ambient temperature x-ray diffraction was carried out for 21:79.
This time the starting material was made of 21:79 glass. Glass was formed by quenching the
21:79 melt, and then grinding it into powder for x-ray. Temperatures at which glass powder was
x-rayed in ºC were: 25, 320, 410, 430, 450, 470, 500, 550, and 585. The general result is tha
21:79 glass stay
inning of the crystallization process. From 410 ˚C to 585 ˚C, peaks are well established, and
very few changes take place. To better understand what happens between 320 ˚C and 410 ˚C
new glass powder was x-rayed at these following temperatures in ºC: 320, 340, 360, 380, 400,
415, and 430. Figures 5.29A and 5.29B show this process in detail. The x-ray pattern at 380 ˚C
was not included because no significant change at thi
u
139
were hase and
8 refers to Zn2Te3O8 phase.
Figure 5.29A 21:79 glass devitrifies as a response to a temperature increase. The appearing peaksmatched with the appropriate PDF cards in 2 theta range between 25-40 degrees. O2 refers to TeO2 p
O
Figure 5.29B A 21:79 glass devitrifies as a response to a temperature increase further up to 430 ˚C. The
appearing peaks were matched with the appropriate PDF cards in 2 theta range between 20 to 60 degrees.
140
To study the final forming phases, an x-ray pattern was collected for the previous devitrified
glass. This pattern was collected at room temperature using CuKα radiation. The results are
shown in Figures 5.30A and 5.30B.
SFigure 5.30A X-ray pattern for 21:79 glass devitrified at 585 ˚C. Data was collected at room temperature.
imulation shows that TeO2 bears 49% while the rest is Zn2Te3O8.
Figure 5.30B X-ray pattern for 21:79 glass devitrifie at 585 ˚C. Data was collected at room temperature.
Simulation shows that TeO2 bear 49% while the rest is Zn2Te3O8. d s
141
Using the lever rule for phase diagrams [7], TeO tage was calculated to be 47.5% and
52.5% is Zn2Te3O8, which is in an agreement with the simulation.
5.2.4 ZnO:TeO2 - 33.3:66.7
This calcined mole percentage was melted at 700 ˚C in a temperature-controlled process,
with a cooling rate of 20 ˚C/hr. Data results obtained from the power curve of this run show the
formation of two phases, one at 665 ˚C and the second one is at 625 ˚C. Microprobe results
show that the material has both Zn2Te3O8 and 20-30% TeO2. X-ray results confirm the formation
of the two phases mentioned above as shown in Figures 5.31A and 5.31B, where the x-ray
pattern was simulated with TeO2 and Zn2Te3O8 theoretical x-ray patterns, using CrystalDiffract.
The simulation shows that there is 18% TeO2 and 82% Zn2Te3O8 with residues of ZnO. At this
composition, no ZnTeO3 was noticed yet.
2 percen
Figure 5.31A X-ray for 33.3:66.7 material melted at 700 ˚C simulated with the TeO2 and Zn2Te3O8 mixture.
142
Figure 5.31B X-ray for 33.3:66.7 material melted at 700 ˚C simulated with the TeO2 and Zn2Te3O8 mixture.
5.2.5 ZnO:TeO - 36.5:63.5
A calcined 36.5:63.5 mixture was melted for 17 hours at 695 ºC then cooled down quickly
(150-200 ºC/hour) to room temperature. A sample was taken from the center of the ingot for
powder x-ray measurement. The resultant x-ray pattern and its simulation are shown in Figure
5.32 and it contains 9.4% TeO2, 89% Zn2Te3O8, and 1.6% ZnTeO3.
2
143
it has 9.4% TeOFigure 5.32 X-ray pattern for 36.5:63.5 material extracted from the center of the ingot. Simulation shows that
5.2.6 Zn
This m elted, pulled, and grown. Figure 5.33 shows a tem
controlled cooling down of 40:60
2, 89% Zn2Te3O8, and 1.6% ZnTeO3.
O:TeO2 - 40:60
ole percentage was m perature-
from melt at 40 ºC/ hour.
Figure 5.33 Phase formation of 40:60 material upon cooling down at 40 degrees/hr.
144
The
This be oled down at a rate between 40 and 50
ºC/
In o
ceramic
four dif 2%
ZnTe5O ce number of 6. Up until the end of
x-ray diffraction library. Figures 5.34 A, 5.34B and
5.34C s
most fo sult
is confi
± 3% o
power curve as a function of time shows three dips, each dip marks a phase formation.
havior was noticed in each 40:60 run that was co
hr.
ne run, where the melt of this mole percentage was pulled, the resulting material had a
-like texture. Scanning electron microprobe shows that the pulled material is formed of
ferent phases: two are major (~37% Zn2Te3O8, ~60%ZnTeO3), and two are minor (~
11, ~1% Zn3TeO6). In the last phase, Te has a valan
2004, ZnTe5O11 had no PDF card in the
how BSE images of the material. The first image is representative, and it shows the two
rming phases in this system. 60% ZnTeO3 and the rest is almost all Zn2Te3O8. This re
rmed using Clemex Vision Analysis software. The margin of error in this measurement is
f each phase.
Figu rea is
ZnTeO3. This image is almost representative.
How wo
other p
of the p
re 5.34A BSE image of 40:60 pulled material. The yellow area is Zn2Te3O8 and the orange a
ever, when focusing the electron beam on certain areas, it was noticed that there are t
hases, ZnTe5O11 and Zn3TeO6. Figures 34B and 34C show these phases. The investigation
resence of these two phases was conducted more than once. The general phase
145
per t
collecti
cen age summation of the forming oxides and the atoms numbers are given in table 5.5. Data
on and analysis were done six times for Zn3TeO6, and twice for ZnTe5O11.
Figure 5.34B BSE image of 40:60 pulled material. he yellow area is ZnTe5O11 and the orange area is
Zn2Te3O8. This image is not representative.
T
Figure 5.34C BSE image of 40:60 pulled material. Th low area is Zn2Te3O8, the orange area is Zn3TeO6,
and the white area is ZnTe5O11. This image is not representative.
e yel
146
Table 5.5 Scanning microprobe analysis for 40:60 pulled material.
TeO2 inclusions, cooling at 40K/hr with Bridgman. Crystals were
chosen/ SEM and XRD
153
Table 5.7 Percentage of each phase for some compositions found by CrystalDiffract 1.3. The margin of error or each phase is ± 5%. esulting materials found by CrystalDiffract % at the isotherm line in the phase diagram, where
W. Journal of Solid State Chemistry, 143: 246-253,
ering Materials
Sciences, 25, No 3: 329 – 331, 1972. [2] H. Bürger, K. Kneipp, H. Hobert, and W. Vogel, Journal of N- 142, 1992. [3] A. Nukui, T. Taniguchi, M. Miyata, Journal of Non-Crystallin2001. [4
[5] K. Hanke, V. Kupcik, and
[6] C.R. Feger, G.L. Schimek, and Kolis, J.
1999.
[7] V. John, Introduction to Engine , 3rd Edition, Antony Rowe Limited,
Great Britain, pp. 144 – 147, 1992.
iams, and R. S. Mitchell, Canadian Meneralogist, 7: 443 – 452,
. Z. Ding, and Z. Q. Hu, Journal of Materials Research:19(9), 2523
.Vogel, Journal of Non-Crystalline Solids: 151, 134 - (1992).
in, V. Kalem, and M. L Öveçoğlu, Key Engineering Materials:
Appendix 4 ZnTe B6 BO B13B crystal structure and parameters
Figure A4.1 Ball and stick representation of the extended Structure of ZnTeB6BO B13B. Zn atoms are enclosed in
light blue polyhedra. Atom legend and unit cell orientation are shown at bottom left.
186
Figure A4.2 Ball and stick representation of the extended Structure of ZnTeB6BO B13B. Zn atoms are enclosed in light blue polyhedra. Atom legend and unit cell orientation are shown at bottom left.
187
Table A4.1 General equivalent positions. +x +y +z -y +x-y +z -x+y -x +z -x -y -z +y -x+y -z +x-y +x -z Table A4.2 Listing of atomic coordinates for first unit cell (Total of 120 atoms in the complete unit cell). +---------------------------------------------------------------------------------------+ | Fractional Coordinates Orthogonal Coordinates | | Label Elmt x y z xor yor zor | +----------------------------------------------------------------------------------------+ | O1 O 0.00000 0.00000 0.08680 0.00090 1.57775 -0.46441 | | O1 O 0.66667 0.33333 0.42013 1.15403 6.01716 -7.74779 | | O1 O 0.33333 0.66667 0.75347 -4.38268 12.60735 -7.73717 | | O1 O 0.00000 0.00000 0.91320 0.00943 16.59912 -4.88591 | | O1 O 0.66667 0.33333 0.24653 1.15224 2.86166 -6.81897 | | O1 O 0.33333 0.66667 0.57987 -4.38447 9.45184 -6.80836 | | O2 O 0.20380 0.11550 0.19260 0.21839 2.99819 -2.73780 | | O2 O 0.87047 0.44883 0.52593 1.37153 7.43760 -10.02118 | | O2 O 0.53713 0.78217 0.85927 -4.16518 14.02779 -10.01057 | | O2 O 0.88450 0.08830 0.19260 5.04228 1.54932 -7.65083 | | O2 O 0.55117 0.42163 0.52593 -0.49443 8.13950 -7.64021 | | O2 O 0.21783 0.75497 0.85927 -6.03114 14.72968 -7.62959 | | O2 O 0.91170 0.79620 0.19260 -1.80563 1.09644 -9.20261 | | O2 O 0.57837 0.12953 0.52593 2.58827 8.24373 -7.28016 | | O2 O 0.24503 0.46287 0.85927 -2.94844 14.83391 -7.26955 | | O2 O 0.79620 0.88450 0.80740 -3.44883 12.47081 -11.81834 | | O2 O 0.46287 0.21783 0.14073 0.93474 1.44122 -4.54558 | | O2 O 0.12953 0.55117 0.47407 -4.60197 8.03140 -4.53496 | | O2 O 0.11550 0.91170 0.80740 -8.27272 13.91969 -6.90532 | | O2 O 0.78217 0.24503 0.14073 2.80070 0.73932 -6.92655 | | O2 O 0.44883 0.57837 0.47407 -2.73601 7.32950 -6.91593 | | O2 O 0.08830 0.20380 0.80740 -1.42481 14.37256 -5.35354 | | O2 O 0.75497 0.53713 0.14073 -0.28200 0.63509 -7.28660 | | O2 O 0.42163 0.87047 0.47407 -5.81871 7.22528 -7.27598 | | O3 O 0.25150 0.25090 0.05110 -0.80857 0.24814 -2.58752 | | O3 O 0.91817 0.58423 0.38443 0.34457 4.68755 -9.87090 | | O3 O 0.58483 0.91757 0.71777 -5.19214 11.27773 -9.86028 | | O3 O 0.74910 0.00060 0.05110 5.00593 -0.68264 -5.73848 | | O3 O 0.41577 0.33393 0.38443 -0.53078 5.90754 -5.72786 | | O3 O 0.08243 0.66727 0.71777 -6.06749 12.49772 -5.71725 | | O3 O 0.99940 0.74850 0.05110 -0.74670 -1.63763 -8.99402 | | O3 O 0.66607 0.08183 0.38443 3.64720 5.50965 -7.07158 | | O3 O 0.33273 0.41517 0.71777 -1.88951 12.09983 -7.06096 | | O3 O 0.74850 0.74910 0.94890 -2.42187 15.22086 -11.96863 | | O3 O 0.41517 0.08243 0.28223 1.96170 4.19127 -4.69586 |
188
| O3 O 0.08183 0.41577 0.61557 -3.57501 10.78145 -4.68524 | | O3 O 0.25090 0.99940 0.94890 -8.23637 16.15164 -8.81766 | | O3 O 0.91757 0.33273 0.28223 2.83705 2.97128 -8.83890 | | O3 O 0.58423 0.66607 0.61557 -2.69966 9.56146 -8.82828 | | O3 O 0.00060 0.25150 0.94890 -2.48374 17.10664 -5.56212 | | O3 O 0.66727 0.58483 0.28223 -1.34093 3.36917 -7.49518 | | O3 O 0.33393 0.91817 0.61557 -6.87764 9.95935 -7.48456 | | O4 O 0.20370 0.48260 0.09770 -3.42878 1.06891 -2.93116 | | O4 O 0.87037 0.81593 0.43103 -2.27564 5.50832 -10.21454 | | O4 O 0.53703 0.14927 0.76437 2.11825 12.65561 -8.29210 | | O4 O 0.51740 0.72110 0.09770 -3.69863 0.26135 -5.67526 | | O4 O 0.18407 0.05443 0.43103 0.69527 7.40863 -3.75282 | | O4 O 0.85073 0.38777 0.76437 1.84841 11.84804 -11.03620 | | O4 O 0.27890 0.79630 0.09770 -6.04094 0.73241 -4.07941 | | O4 O 0.94557 0.12963 0.43103 5.04281 5.72892 -9.45096 | | O4 O 0.61223 0.46297 0.76437 -0.49390 12.31911 -9.44035 | | O4 O 0.79630 0.51740 0.90230 0.19834 14.40009 -11.62498 | | O4 O 0.46297 0.85073 0.23563 -5.34869 2.81340 -6.26404 | | O4 O 0.12963 0.18407 0.56897 -0.95479 9.96068 -4.34160 | | O4 O 0.48260 0.27890 0.90230 0.46819 15.20766 -8.88089 | | O4 O 0.14927 0.61223 0.23563 -5.07885 3.62096 -3.51995 | | O4 O 0.81593 0.94557 0.56897 -3.92571 8.06038 -10.80333 | | O4 O 0.72110 0.20370 0.90230 2.81050 14.73659 -10.47674 | | O4 O 0.38777 0.53703 0.23563 -2.73654 3.14990 -5.11580 | | O4 O 0.05443 0.87037 0.56897 -8.27325 9.74008 -5.10518 | | O5 O 0.38430 0.52290 0.97990 -2.61169 16.69367 -9.04556 | | O5 O 0.05097 0.85623 0.31323 -8.15872 5.10698 -3.68462 | | O5 O 0.71763 0.18957 0.64657 2.92502 10.10349 -9.05617 | | O5 O 0.47710 0.86140 0.97990 -5.35238 16.30550 -10.36959 | | O5 O 0.14377 0.19473 0.31323 -0.96881 5.27591 -3.09683 | | O5 O 0.81043 0.52807 0.64657 0.18433 9.71532 -10.38021 | | O5 O 0.13860 0.61570 0.97990 -5.17695 17.17042 -7.43084 | | O5 O 0.80527 0.94903 0.31323 -4.03413 3.43295 -9.36390 | | O5 O 0.47193 0.28237 0.64657 0.35976 10.58024 -7.44146 | | O5 O 0.61570 0.47710 0.02010 -0.61875 -1.22467 -5.51059 | | O5 O 0.28237 0.81043 0.35343 -6.15546 5.36551 -5.49997 | | O5 O 0.94903 0.14377 0.68677 4.92829 10.36203 -10.87153 | | O5 O 0.52290 0.13860 0.02010 2.12195 -0.83650 -4.18655 | | O5 O 0.18957 0.47193 0.35343 -3.41476 5.75368 -4.17593 | | O5 O 0.85623 0.80527 0.68677 -2.26163 10.19310 -11.45932 | | O5 O 0.86140 0.38430 0.02010 1.94651 -1.70141 -7.12531 | | O5 O 0.52807 0.71763 0.35343 -3.59020 4.88877 -7.11469 | | O5 O 0.19473 0.05097 0.68677 0.80369 12.03605 -5.19225 | | Te1 Te 0.23760 0.08143 0.09533 0.78184 1.17641 -2.39878 | | Te1 Te 0.90427 0.41476 0.42866 1.93498 5.61582 -9.68216 | | Te1 Te 0.57093 0.74810 0.76200 -3.60173 12.20601 -9.67154 |