as illliHHlijm

Ht 293 02074 1009

0:9 LIL} '9“th
PAIChEC’C 923:?
University

                                                                                                   

 

 

 

 

This is to certify that the

dissertation entitled

Studies on the Reactivity of Coinage Metals and
Rare Earth Metals in Alkali Metal/Polytelluride Fluxes

presented by

Rhonda Renee Patschke

has been accepted towards fulfillment
of the requirements for

Ph . D . Chemis try

degree in

 

 

 

Major professor

9&12fl1/tt

MSU is an Affirmative Action/Equal Opportunity Institution 0. 12771

 

 

PMCE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11m chIRCJDatODII.w5-p.14

 

 

STUDIES ON THE REACTIVITY OF COINAGE METALS AND RARE
EARTH METALS IN ALKALI METAL/POLYTELLURIDE F LUXES

By

Rhonda Reneé Patschke

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1999

ABSTRACT

STUDIES ON THE REACTIVITY OF COINAGE METALS AND RARE
EARTH METALS IN ALKALI METAL/POLYTELLURIDE F LUXES

By

Rhonda Renee Patschke

Over the past decade, the polychalcogenide flux method has become an

established technique for discovering new solid state compounds. The advantage
to using molten fluxes is that they allow the reaction system to choose its own
route (either kinetic or thermodynamic) without forcing it to a certain
stoichiometry or structure type. The materials discovered by this method can be
used for a wide variety of applications, including batteries, lasers, non-linear
optics, photovoltaics, composites, and thermoelectrics. In order to maximize the
probability that the compounds formed will possess new structure types, a
quaternary system was explored in which two metals with very different
coordination preferences, namely a coinage metal and a rare earth metal, were
reacted in an alkali metal/polychalcogenide flux. Explorations in this system with
sulfur and selenium proved to be successful, and so we decided to expand this
Chemistry into the polytelluride system. As a result, several new compounds were

discovered and most notably, the structure types formed were very different than

It. ‘.\I

.4 n
b ’4‘
in .v. .Lq

 

g‘...
H...

 

o ~.
5 ,_

1

‘5

I s
,- s.
.‘u _

"l

those previously found in the sulfur and selenium systems. Many of the
compounds contain stacking layers that can be described as a square Te net. These
Te nets have a propensity to distort, which gives rise to subtle crystallographic
superstructures.

In this dissertation, the synthesis, structure, and physicochemical properties
of many new phases will be reported. In systems where only a rare earth metal
reacted in a molten A2Tex flux, the family of compounds ALl’l3T63 was discovered
whose members include CsCe3Teg, RbCC3T63, KCC3T83, and KNd3Te3. Several
quaternary phases of the type AwaLnyTez were also discovered by reacting both
a coinage metal and a rare earth metal in a molten A2Tex flux, including
CSCDUTC3, RbCuUTe3, KCuUTe3, KCuCeTe4, RbCuCeTe4, NaugAguCeTm,
K25Ag45Ce2Te9, K2,5Ag4,5La2Te9, K2Ag3CeTe4, szCu3CeTe5, KCquuTe4,
NamAgnEuTen and K0,65Ag2Eu1.35Te4. In the case where a flux was not used,
the ternary phase CUXUTC3 (x = 0.25 and 0.33) was discovered. Finally, the cage
compounds A2MCugTe1o (KzBaCugTem szBaCugTem, CszBaCugTem, and

szEllCllgTCm) were investigated for their promising thermoelectric properties.

 

 

 

 

 

ACKNOWLEDGEMENTS

First of all, I wish to thank the people who inspired me early on in my
career. Without them, I might have fallen through the cracks and then certainly
would not have made it here today. These people include Dr. Raghu Menon, Dr.
Emel Yakali, Dr. Ken Koehler, Mrs. Ann Spector, Dr. Bruce Ault, and Dr.
Barbara Stout.

As far as my time at Michigan State University is concerned, I owe a great
deal to my advisor, Professor Mercouri Kanatzidis. Over the past 5 years, he has
given me a tremendous amount of encouragement, support, and guidance. He has
helped me to strive to be the scientist I am today and for that, I will always be
grateful. I consider him to be a role model for many reasons. Mostly, however, I
admire how he has managed to truly preserve his basic love for science over the
years. He demonstrates this every day of his life. I only hope that I am able to do
the same because I, also, want to wake up every day and be excited to go to work.

Next, I would like to thank the people with whom I have collaborated.
Without them, I would not have been able to accomplish the quality of work
presented throughout this dissertation. These collaborations have not only taught
me how to work as a team, but also that I don’t necessarily need to limit myself to

questions within my research that only I can answer. These people include Prof.

Carl Kannewurf, Dr. Jon Schindler, Mr. Paul Brazis, DL Victor Young, Prof.

iv

 

 

 

 

 

 

 

 

Sander van Smallen, Prof. Michel Evain, Prof. Terry Tritt, Mr. Nathan Lowhom,
and Dr. George Nolas.

In addition, I would like to thank the staff at Michigan State University who
helped me with the techniques I needed for my research. These people include Dr.
John Heckman, Dr. Reza Loloee and Dr. Jerry, Cowen, Dr. Stan Flegler, Dr. Rui
Wang, and Dr. Don Ward.

Furthermore, I would like to thank the members of the Kanatzidis group,
past and present, for all of their support. These are the people who cheered me up
when my experiments weren’t working and celebrated with me when they were.

Last but not least, I would like to recognize my fiancée, Paul Willigan. I
cannot even begin to explain how much love and support he has given me over the
years. Even though he was 300 miles away, I needed him every step of the way
and am looking forward to spending the rest of my life “geographically” closer to
him.

Finally, the National Science Foundation and the Center for Fundamental

Materials Research is gratefully acknowledged for financial support.

Np
-,_

ii")
m...‘

. ’0'

II b".

in“.-

{‘I'"Y

p

TABLE OF CONTENTS

Egg
LIST OF TABLES ......................................................................... xiii
LIST OF FIGURES ........................................................................ xix
Chapter 1: The Rationale for Combining Coinage Metals with
Rare Earth Metals in Polytelluride Fluxes ................................................. l
A. Introduction ................................................................... 2
B. Nature of the Polychalcogenide Flux ...................................... 3
C. Synthetic Approach .......................................................... 7
D. Review of Quaternary A/M/Ln/Q Phases ................................. 8
E. Te Net Distortions ........................................................... 20
ChaPter 2: Reactions of Rare Earth Metals in Molten Alkali Metal/
Polytelluride Fluxes: Discovery of the ALn3Te3 Family (A = K, Rb, CS;
Ln = Ce, Nd) ................................................................................... 33
A- Introduction ................................................................. 34
13. Experimental Section ....................................................... 35
1. Reagents ............................................................. 35
Potassium Telluride, KzTe ................................... 35
Rubidium Telluride, szTe .................................. 36
Cesium Telluride, CszTe ..................................... 37
2. Synthesis ............................................................. 37
CsCe3Tes (I) ................................................... 37
RbCe3Teg (II) .................................................. 39
KCe3Te3 (HI) .................................................. 40

 

Q
1“»
3".

‘N w.

KNd3TCg (IV) .................................................. 40

3. Physical Measurements ............................................ 46

C. Results and Discussion ......................................................... 53
Structure Description .......................................... 5 3

Transmission Electron Microscopy ......................... 66

Magnetic Susceptibility Measurements. . . . .. . . . . . ....69
Charge Transport Properties ................................. 71

D. Conclusions ..................................................................... 75

Chapter 3: Structure and Properties of ACuUTe3 (A = Cs, Rb, K): A

Comparison with KCuUSe3 ................................................................ 80
A. Introduction .................................................................. 81
B. Experimental Section ........................................................ 84

1. Reagents ............................................................. 84

Cesium Telluride, CszTe ...................................... 84
Rubidium Telluride, szTe .................................. 84

Potassium Telluride, KzTe ................................... 84

2. Synthesis ............................................................ 84
ACuUTe3 (A = Cs, Rb, K) (I-III) ........................... 84

3. Physical Measurements ........................................... 90
C. Results and Discussion ......................................................... 93
Structure Description .......................................... 93

Magnetic Susceptibility Measurements ................... 97
Charge Transport Properties ................................. 99

Infi'ared Spectroscopy ...................................... 101

Cation Effect on ACuUTe3 ................................. 103

D. Conclusions .................................................................... 105

vii

 

Chapter 4: Novel Polytelluride Compounds Containing Distorted

Nets of Tellurium .......................................................................... 108
A. Introduction ................................................................... 109
B. Experimental Section ......................................................... 111

1. Reagents ............................................................. 111
Sodium Telluride, NazTe ................................... 112
Potassium Telluride, KzTe ................................. 113
Rubidium Telluride, szTe ................................ 113

2. Synthesis ............................................................ 113
KCuCeTe4 (I) ................................................ 113
RbCuCeTe, (II) .............................................. 114
Nao_8Ag.,2CeTe4 (III) ........................................ 114
K2,5Ag4,5Ce2Te9 (IV) ........................................ 115
K2_5Ag4_5La2Te9 (V) ........................................ 1 16
Cuo,“EuTe2 (V 1) ............................................ 1 17
KCquuTe4 (VII) ............................................ 118
Nao_2Ag2,8EuTe4 (VIII) ..................................... 119
Ko_55Ag2Eu,,35Te4 (IX) ...................................... 119

3. Physical Measurements ........................................... 130

C. Results and Discussion ..................................................... 135

1. AXM(2_X)CeTe4 (I, II, HI) ........................................... 135
Structure Description ....................................... 135
Transmission Electron Microscopy ....................... 146
Magnetic Susceptibility Measurements. .. . . . . . . . . 152
Infrared Spectroscopy ...................................... 154
Charge Transport Properties ................................ 155

2. K2_5Ag4,5Ln2Te9 (Ln=Ce,La) (IV, V) ............................. 157
Structure Description ........................................ 15 7
Transmission Electron Microscopy ....................... 165

viii

Superstructure Determination ............................. 168

Superstructure Description ................................. 173
Magnetic Susceptibility and Infrared

Spectroscopy ............................................... 185

Charge Transport Properties ................................ 187

3. CucMEuTez (VI) ................................................... 191

Structure Description ....................................... 191

4. AxM(3_x)EuTe4 (VII, VIII) ......................................... 195

Structure Description ......................................... 195

Transmission Electron Microscopy ..... ' .................. 203

Magnetic Susceptibility Measurements ................. 207

Infi’ared Spectroscopy ....................................... 209

Charge Transport Properties ............................... 210

5. Ko.65Ag2Eu1,35Te4 (VII) ............................................ 212

Structure Description ........................................ 212
Magnetic Susceptibility and Infrared

Spectroscopy ................................................ 222

Charge Transport Properties ................................ 224

D. Conclusions .................................................................... 226

Chapter 5: Novel Quaternary Polytelluride Compounds Without

Te Nets ...................................................................................... 231
A Introduction ................................................................... 232

3° Experimental Section ......................................................... 233

1. Reagents ............................................................ 233

Potassium Telluride, KzTe ............................... 234

Rubidium Telluride, szTe ............................. 234

2. Synthesis ............................................................ 234

K2Ag3CeTe4 (I) ............................................... 234

 

 

 

 

RbZCu3CeTe5 (H) ............................................ 235

3. Physical Measurements .......................................... 241

C. Results and Discussion ....................................................... 244
Structure Description of K2Ag3CeTe4 (I) ................ 244

Structure Description of szCugCeTes (II) .............. 245

Ion-Exchange Properties of K2Ag3CeTe4 (I) ............259
Magnetic Susceptibility and Infrared

Spectroscopy ................................................ 262
Charge Transport Properties ............................... 266
D. Conclusions ................................................................... 269

Chapter 6 CuxUTeg, (x = 0.25 and 0.33): Stabilization of UTe3 in

the ZrSe3 Structure Type via Copper Insertion ......................................... 272
A. Introduction ................................................................... 273
B. Experimental .................................................................. 274

1. Reagents ............................................................ 274

2. Synthesis ........................................................... 274
CuxUTe3 (x = 0.25 and 0.33) .............................. 274

3. Physical Measurements .......................................... 275

C. Results and Discussion ..................................................... 278
Structure Description ....................................... 278

0t— vs B—type UTe3 ......................................... 285

Superstructure ................................................ 293

Transmission Electron Microscopy ....................... 293

Charge Transport Properties ............................... 299

D. Conclusions .................................................................. 303

 

 

Synthesis and Thermoelectric Studies of the Cage

Chapter 7
Compounds, AzMCusTcio (A = K, Rb, Cs; M = Ba, Eu) ............................. 307
A. Introduction .................................................................. 308
B. Experimental ................................................................ 309
1. Reagents ............................................................ 309
Potassium Telluride, K2Te ................................. 310
Rubidium Telluride, szTe ................................ 310
Cesium Telluride, CszTe .................................. 310
Europium Telluride, EuTe ................................. 310
2. Synthesis ........................................................... 311
A2MC113T610 (A = K, Rb, Cs; M = Ba, Eu) ............. 311
3. Physical Measurements .......................................... 317
C. Results and Discussion ..................................................... 321
Structure Description ........................................ 321
Charge Transport Properties ............................... 349
Infrared Spectroscopy ..................................... 338
Heat Capacity ............................................... 340
Raman Spectroscopy ....................................... 345
Magnetic Susceptibility .................................... 346
Thermal Analysis ........................................... 349
D. Conclusions ................................................................ 352

xi

 

a...

 

 

LIST OF TABLES

Page

Table 1.1 Melting Points for Some Known Alkali Metal/

Polychalcogenide (AzQQ Species .......................................... 5
Table 2.1 Calculated and Observed X-ray Powder Diffraction

Pattern for CSCC3T68 (I) .................................................... 42
Table 2.2 Calculated and Observed X-ray Powder Diffraction

Pattern for RbCe3Te3 (II) .................................................. 43
Table 2.3 Calculated and Observed X-ray Powder Diffraction

Pattern for KCC3T€3 (III) .................................................... 44
Table 2.4 Calculated and Observed X-ray Powder Diffraction

Pattern for KngTeg (IV) .................................................. 45
Table 2.5 Crystallographic Data for ALn3Te8 (A = Cs, Rb, K;

Ln = Ce, Nd) ................................................................. 51
Table 2.6 Emotional Atomic Coordinates and Equivalent Isotropic

Displacement Parameters (Beq) for ALH3TCg (A = Cs, Rb,

K; Ln = Ce, Nd) with Estimated Standard Deviations in

Parentheses .................................................................. 58
Table 2.7 Anisotropic Displacement Parameters (A) for ALn3Te8

(A = Cs, Rb, K; Ln = Ce, Nd) with Standard Deviations

in Parentheses .............................................................. 60
Table 2.8 Selected Distances (A) and Bond Angles (deg) for

CsCe3Te8with Standard Deviations in Parentheses. . . . . . . . . . . . . . .......62
Table 2.9 Selected Distances (A) and Bond Angles (deg) for

RbCe3Te8with Standard Deviations in Parentheses.....................63
Table 2.10 Selected Distances (A) and Bond Angles (deg) for

KCe3Te8with Standard Deviations in Parentheses.......... . . .......64

xii

.
'I

.
.g'v ..,

 

Table 2.11

Table 3.1

Table 3.2

Table 3.3

Table 3.4

Table 4.1

Table 4.2

Table 4.3

Table 4.4

Table 4.5

Table 4.6

Table 4.7

Table 4.8

Table 4.9

Selected Distances (A) and Bond Angles (deg) for

IQQd3Te8with Standard Deviations in Parentheses.....................65
Calculated and Observed X-ray Powder Diffraction

Pattern for CsCuUTe3 (III) ................................................ 86
Calculated and Observed X-ray Powder Diffiaction

Pattern for RbCuUTe; (II) ................................................. 88
Calculated and Observed X-ray Powder Diffraction

Pattern for KCuUTe3 (III) ................................................. 89
Unit Cell Parameters for ACuUTe3 (A = Cs, Rb, K).......... ........92
Calculated and Observed X-ray Powder Diffraction

Pattern for KCuCeTe4 (I) ................................................ 121
Calculated and Observed X—ray Powder Diffraction

Pattern for NaMAngeTe4 (III) ....................................... 122
Calculated and Observed X-ray Powder Diffraction

Pattern for KlsAgrsCe/zTeg (IV) (based on superstructure). . . . . 123
Calculated and Observed X-ray Powder Diffraction

Pattern for K2.4Ag4,6La2Te9 (V) (based on superstructure) ......... 124
Calculated and Observed X-ray Powder Diffraction

Pattern for CumgéEuTe2 (VI) ............................................. 126
Calculated and Observed X-ray Powder Diffraction

Pattern for KCquuTe4 (VII) ............................................ 127
Calculated and Observed X-ray Powder Diffraction

Calculated and Observed X-ray Powder Diffraction

Pattern for Kofi65Ag2EuL35TC4 (IX) ...................................... 129
Crystallographic Data for KCuCeTe4 (I), RbCuCeTe4

(II), and NaogAngeTe4 (III) .......................................... I32

xiii

Table 4.10

Table 4.1 1

Table 4.12

Table 4.13

Table 4.14

Table 4.15

Table 4.16

Table 4.17

Table 4.18

Table 4.19

Table 4.20

Crystallographic Data for K2_5Ag4,5Ce2Teo (IV),

K25Ag45132T89 (V), and CUQMEUTCZ (VI) ........................... 133

Crystallographic Data for KCquuTe4 (VII),

Nao,2Ag2.sEuTe4 (VIII), and Ko_65Ag2Eu1_35Te4 (IX) ...............

Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Um) for KCuCeTe4 (I),
RbCuCeTe4 (II), and NaagAguCeTe.‘ (III) with

Estimated Standard Deviations in Parentheses .....................

Anisotropic Displacement Parameters (A) for KCuCeTe4
(I), RbCuCeTe4 (II), and Nao,gAg,,2CeTe4 (III) with

Standard Deviations in Parentheses ..................................

Selected Distances (A) and Bond Angles (deg) for

KCuCeTe4 (I) with Standard Deviations in Parentheses. . . . . . .

Selected Distances (A) and Bond Angles (deg) for

RbCuCeTe4 (II) with Standard Deviations in Parentheses ........

Selected Distances (A) and Bond Angles (deg) for
NaonAgLZCeTe, (HI) with Standard Deviations in

...134

...140

..141

142

..143

Parentheses ................................................................ 144

Fractional Atomic Coordinates and Equivalent Isotropic

Displacement Parameters (ch) for K2,5Ag4,5Ce2Teo (IV)
and K2.5Ag4,5LazTe9 (V) with Estimated Standard Deviations

in Parentheses .............................................................. 161

Anisotropic Displacement Parameters (A) for
K2,5Ag4_5Ce2Te9 (IV) and K2.5Ag4.5L32TC9 (V) With

Standard Deviations in Parentheses ...................................

Selected Distances (A) and Bond Angles (deg) for
K2,5Ag4,5CezTe9 (IV) with Standard Deviations in

..162

Parentheses ................................................................ 163

Selected Distances (A) and Bond Angles (deg) for
K2.5Ag4,51132T69 (V) with Standard Deviations in

Parentheses ................................................................ 164

xiv

 

Table 4.21

Table 4.22

Table 4.23

Table 4.24

Table 4.25

Table 4.26

Table 4.27

Table 4.28

Table 4.29

Table 4.30

Table 4.31

Crystallographic Data for the “la x 3b” superstructures

0f K2_5Ag4_5C62T69 (IV) and K2,5Ag4_5La2Te9 (V) ..................... 172

Emotional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (UN) for the “1am x 3b,,1, ”
superstructure of K2,5Ag4,5Ce2Te9 (IV) with Estimated

Standard Deviations in Parentheses .................................... 177

Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Ueq) for the “lamb x 3b,“, ”
superstructure of K2,5Ag4.5La2Teo (V) with Estimated

Standard Deviations in Parentheses ..................................... 178

Anisotropic Displacement Parameters (A) for the
“lamb x 3b,“, ” superstructure of K2_5Ag4,5Ce2Teo (IV)

with Standard Deviations in Parentheses ............................... 179

Anisotropic Displacement Parameters (A) for the
“lamb x 3b,”, ” superstructure of K2_5Ag4,5La2Te9 (V)
with Standard Deviations in Parentheses .....................

Selected Distances (A) and Bond Angles (deg) for the
“1am, x 3b,”, ” superstructures of K2_5Ag4,5Ce2Te9 (IV)
with Standard Deviations in Parentheses .....................

Selected Distances (A) and Bond Angles (deg) for the
“lam x 3b,,b ” superstructures of K2,5Ag4,51a2Te9 (IV)
with Standard Deviations in Parentheses .....................

Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Ueq) for Cu0_66EuTe2 (IV) with

......... 180

......... 181

......... 183

Estimated Standard Deviations in Parentheses ........................ 193

Anisotropic Displacement Parameters (A) for Cuo_6(,EuTe2

(IV) With Standard Deviations in Parentheses ......................... 193

Selected Distances (A) and Bond Angles (deg) for
Cuo_66EuTe2 (IV) with Standard Deviations in Parentheses

Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Ueq) for KCquuTe4 (VII)
and Na0_2Ag2,gEuTe4 (VIII) with Estimated Standard

......... 194

Deviations in Parentheses ................................................ 199

 

Table 4.32

Table 4.33

Table 4.34

Table 4.35

Table 4.36

Table 4.37

Table 5.1

Table 5.2

Table 5.3

Table 5.4

Table 5.5

Table 5.6

Anisotropic Displacement Parameters (A) for KCquuTe4
(VII) and Nao.2Ag2,3EuTe4 (VIII) with Standard Deviations
in Parentheses ............................................................. 200

Selected Distances (A) and Bond Angles (deg) for
KCquuTe4 (VII) with Standard Deviations in Parentheses ......... 201

Selected Distances (A) and Bond Angles (deg) for
Nao,2Ag2_3EuTe4 (VIII) with Standard Deviations in
Parentheses ................................................................ 202

Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (8,.) for K0,65Ag2Eu,_35Te4 (DO
with Estimated Standard Deviations in Parentheses .................. 217

Anisotropic Displacement Parameters (A) for
Ko_65Ag2EU1.35TC4 (IX) With Standard Deviations in
Parentheses ................................................................ 218

Selected Distances (A) and Bond Angles (deg) for
K065Ag2Eu135Te4 (IX) with Standard Deviations in
Parentheses ................................................................. 219

Calculated and Observed X-ray Powder Diffi’action
Pattern for KzAg3CCTC4 (I) ............................................... 237

Calculated and Observed X-ray Powder Diffiaction
Pattern for szCu3CeTe5 (H) ............................................. 239

Crystallographic Data for K2Ag3CeTe4 (I) and
szCu3CeTe5 (H) .......................................................... 243

Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Ueq) for KzAg3CeTe4 with
Estimated Standard Deviations in Parentheses ........................ 250

Anisotropic Displacement Parameters (A) for
KzAg3CeTe4 with Standard Deviations in Parentheses .............. 251

Selected Distances (A) and Bond Angles (deg) for
K2Ag3CeTe4 with Standard Deviations in Parentheses .............. 252

xvi

\~\

 

Table 5.7

Table 5.8

Table 5.9

Table 6.1

Table 6.2

Table 6.3

Table 6.4

Table 6.5

Table 7.1

Table 7.2

Table 7.3

Table 7.4

Table 7.5

Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Ueq) for szCu3CeTe5 with

Estimated Standard Deviations in Parentheses ..................

Anisotropic Displacement Parameters (A) for

Rb2Cu3CeTe5 with Standard Deviations in Parentheses. . . . . . . .

Selected Distances (A) and Bond Angles (deg) for

RbZCu3CeTe5 with Standard Deviations in Parentheses. . . . . . .

Crystallographic Data for CuxUTe3 (x = 0.25 and 0.33) .......

Fractional Atomic Coordinates, Equivalent Isotropic
Displacement Parameters (Ueq), and occupancies for
Cllo_25UTC3 (Crystal #3) with Estimated Standard

Deviations in Parentheses ...........................................

Anisotropic Displacement Parameters (A2) for
Cuo,25UTe3 (Crystal #3) with Estimated Standard

Deviations in Parentheses ..........................................

Selected Distances (A) and Bond Angles (deg) for
Cuo_25UTe3 (Crystal #3) with Standard Deviations in

...... 256

.257

.258

277

...... 281

...... 281

Parentheses ................................................................ 282

Relative Stability of the UTe3 structure types as a

function of amount of tellurium added ...........................

Calculated and Observed X-ray Powder Diffraction

Pattern for K2B3CUgTClo (I) ..........................................

Calculated and Observed X-ray Powder Diffiaction

Pattern for szBaCugTelo (II) .......................................

Calculated and Observed X-ray Powder Diffraction

Pattern for CszBaCu3Telo (HI) ......................................

Calculated and Observed X-ray Powder Diffraction

Pattern for szEuCugTem (IV) .....................................

Crystallographic Data for AzBaCugTelo (A = K, Rb, Cs)...

xvii

289

...... 313

...... 314

...... 315

316

....320

Fifi
I

..

...n

«u.
It.

 

Table 7.6

Table 7.7

Table 7.8

Table 7.9

Fractional Atomic Coordinates and Equivalent Isotropic

Displacement Parameters (Ueq) for szBaCugTem with
Estimated Standard Deviations in Parentheses ........................ 327

Anisotropic Displacement Parameters (A) for szBaCllgTClo
with Standard Deviations in Parentheses ............................... 327

Selected Distances (A) and Bond Angles (deg) for
szBaCusTelo with Standard Deviations in Parentheses ............ 328

Room temperature values for the electrical conductivity,
thermopower, and heat capacity of AzMCllgTelo (A = K,

Rb, Cs; M = Ba, Eu) and the Debye temperatures and 7

values derived from the heat capacity .................................. 344

xviii

Figure 1.1

Figure 1.2

Figure 1.3

Figure 1.4

Figure 1.5

Figure 2.1

Figure 2.2

Figure 2.3

Figure 2.4

Figure 2.5

LIST OF FIGURES

Extended structure of ACuLn2Q6 as seen down

the b-axis ..............................................................

Extended structure of K2Cu2CeS4 as seen down

the b—axis ..............................................................

(A) Cmcm structure type of AMLnQ3 (e.g.; KCuUSe3)
(B) ana (I) structure type of AMLnQ3 (e. g.; N3CUTIS3)
(C) C2/m structure type of AMLnQ; (e.g.; BaAgErS3) and

(D) ana (11) structure type of AMLnQ; (e.g.; BaCuIaS3). . . . .

Extended structure of KCquS4 as seen down the a-axis ............ l8

ORTEP representation of the extended structure of

K6C1112U2815 ......................................................

ORTEP representation of the structure of ALn3Te3 as

seen parallel to the anionic layer ...................................

A fragment of CsCe3Te3 showing the coordination

environment of the Ce atoms .......................................

View of the Te “net” of CsCe3Te8 showing the Te32‘

units and the infinite zigzag (Tezz')I1 chains .......................

(A) Selected are electron diffraction pattern of KNd3Te3
with the beam perpendicular to the layers ([001] direction)
(B) Densitometric intensity scan along the b -axis of the

electron diffraction pattern ..........................................

Inverse molar magnetic susceptibility (l/xM) plotted
against temperature (2-300K) for (A) RbCC3TCg, (B)

KCegTeg, and (C) KNd3Teg ..........................................

xix

Page

....... l4

....... 15

16

..... l9

...... 55

....... 56

...... 57

...... 67

...... 7O

 

.“.
u

.
Iv,»

'.
,"‘
“‘..

a

ii‘ll;

w‘A-I.

Figure 2.6 (A) Four probe, electrical conductivity [log 0 (S/cm)]
and (B) Thermopower (S/cm) data plotted against
temperature (K) for room temperature pressed pellets of
CSCC3T63 and RbCe3Te3 and a single crystal of KNd3TCg ............ 73

Figure 2.7 (A) Four probe, electrical conductivity (S/cm) and (B)
Thennopower (uV/K) data plotted against temperature (K)
for room temperature pressed pellets of CeTe3 ........................ 74

Figure 3.1 The four structure types of AMM’Q3 (A = alkali or
alkaline earth metal, M = coinage metal, M’ = Group
IV or rare earth metal, Q = chalcogenide). (A) Cmcm
structure (e.g.; KCquS3), (B) ana (I) structure (e. g.;
NaCuTiS3), C2/m structure (e.g.; BaAgErS3), (D)
ana (11) structure (e. g.; BaCuLaS3) ..................................... 83

Figure 3.2 Polyhedral representation of the one-dimensional
chains built from edge sharing connections of [UT e6]

octahedrons in ACuUTe3 (A = Cs, Rb, K) ............................... 94

Figure 3.3 Polyhedral representation of the two-dimensional corrugated
corrugated layers built from comer sharing connections of the
one-dimensional chains in ACuUTe3 (A = Cs, Rb, K) ................ 94

Figure 3.4 Extended structure of ACuUTe3 (A = Cs, Rb, K)
highlighting how the copper atoms sit in the folds

of the layers .................................................................. 95

Figure 3.5 View perpendicular to a single anionic layer of
ACuUTe; (A = Cs, Rb, K) ................................................ 96

Figure 3.6 Inverse molar magnetic susceptibility (l/xM) plotted
against temperature (2-300K) for (A) CsCuUTe3, (B)

RbCuUTe3, and (C) KCuUTe3 ............................................. 98

Figure 3.7 (A) Variable temperature, four probe electrical
conductivity data for hot pressed pellets of ACuUTe3
(A = Cs, Rb, K). (B) Variable temperature thermopower
data for hot pressed pellets of ACuUTe3 (A = Cs, K) ............... 100

 

 

‘I‘q

‘s

.1

Figure 3.8

Figure 4.1

Figure 4.2

Figure 4.3

Figure 4.4

Figure 4.5

Figure 4.6

Figure 4.7

Diffuse reflectance optical spectra for (A) CsCuUTe3,
(B) RbCuUTe3 and (C) KCuUTe3 (in the Mid-IR

region) ....................................................................... 102

ORTEP representation of the extended structure of
KCuCeTe4 as seen down the b—axis (90% probability

ellipsoids) .................................................................. 138

ORTEP representation (90% probability ellipsoid)

of (A) a view perpendicular to the [CuTe'] layer of

KCuCeTe4, (B) the coordination environment around

Ce in KCuCeTe4, (C) the coordination environment

around K in KCuCeTe4, and (D) a View perpendicular

to the [CCTC3] layer in KCuCeTe4 ...................................... 139

Side by side comparison of the layers in (A) NaCuTe to
those in (B) KCuTe. Both structures are viewed down the

b—axis ....................................................................... 145

(A) Selected area electron diffraction pattern of KCuCeTe4

with the beam perpendicular to the layers ([001] direction)
showing the 2.87am, x 2.87bsub superlattice. (B) Densitometric
intensity scan along the b* axis of the electron diffraction
pattern (boxed area in photograph) showing the ( lkO) family
of reflections. The three reflections from the sublattice of
KCuCeTe4 are indexed. The two weak peaks are mm the

superlattice with bsupa= 2.87bsub ......................................... 148

Cartoon schematic of an electron diffi'action pattern

for (A) a “lamb x 2.87am,” supercell, (B) two “1a,“;J x

287%,,” supercells rotated 90° with respect to one

another and superimposed under the electron beam, and

(C) an apparent “2.87am, x 2.87m,” supercell ....................... 150

View of the Te "net" in KCuCeTe4 showing (A) a “lamb x

2.87b,ub” supercell and (B) a “2.87am x 2.87bsub” supercell.
The crystallographically determined sublattice is shown in

the shaded box for both .................................................. 151

Inverse molar magnetic susceptibility (l/xM) plotted
against temperature (2-300K) for (A) KCuCeTe4, (B)
RbCUCCTC4, and (C) Nao,3Ag1,2CeTe4 .................................. 153

 

Figure 4.8

Figure 4.9

Figure 4.10

Figure 4.11

Figure 4.12

Figure 4.13

Figure 4.14

 

Difi‘use reflectance optical spectra of Na0,3Ag1,2CeTe4

(in the Mid-IR region) .................................................... 154

(A) Four probe, electrical conductivity data of a hot-

pressed pellet of KCuCeTe4 and a room temperature

pressed pellet of Nao_gAg1,2CeTe4 as a function of

temperature. (B) Thermopower data of a hot-pressed

pellet of KCuCeTe4 and a room temperature pressed

pellet of NaogAngCCTC4 as a function of temperature. . . . . . . ......156

ORTEP representation of the extended structure of
K2_5Ag4,5CezTe9 as seen down the b-axis (90% probability

ellipsoids) .................................................................. 159

ORTEP representations of (A) the coordination environment

around Ce in K2,5Ag4,5Ce2Teo (50% probability ellipsoids),

(B) the “Te net” of K2_5Ag4,5Ce2Teo (70% probability

ellipsoids), and (C) the coordination environment around

K in K25Ag45Ce2Te9 (90% probability ellipsoids) ................. 160

(A) Selected area electron diffi’action pattern of

K2,5Ag4,5Ce2Teo with the electron beam perpendicular

to the layers ([001] direction) showing a twinned

3am x 3b,,b domain (i.e.; two lamb x 3bsub supercells

that are rotated 90° with respect to one another and

superimposed). (B) Densitometric intensity scan along

the b*-axis of the electron diffraction pattern of

K2,5Ag4_5Ce2Teo (Fig 4.11 A) (boxed area in photograph)

showing the (h 2 0) family of reflections ............................... 166

ORTEP representation of the “lamb x 3b,,b ”
superstructure of K2_5Ag4,5Ce2Te9 as seen down the

b-axis (75% probability ellipsoids) ...................................... 174

ORTEP representation (50% probability ellipsoids)

of (A) The [K1,5Ag45Te3] layer of the lamb x 3bsub
superstructure of K2,5Ag4_5Ce2Te9, and (B) a fragment

of the [CeTe30'5] layer of the lamb x 3bsub superstructure
of K2,5Ag4_5Ce2Te9 highlighting the particular coordination

enviroment of Ce .......................................................... 175

xxii

1“.‘
M.‘
~ Q
.g‘-"l
K..
‘1.

>-.

1
a.,
.

View of the Te "ne " in (A) the substructure of
K25Ag45CezTe9 and (B) the lamb x 3bsub superstructure
of K2,5Ag4,5Ce2Teo ......................................................... 176

Figure 4.15

(A) Inverse molar magnetic susceptibility (l/xM) plotted

against temperature (2-300K) for K2_5Ag4_5Ce2Te9. (B)

Diffuse reflectance optical spectra of K2,5Ag4_5Ce2Te9 (in

the Mid-IR region) ......................................................... 186

Figure 4.16

(A) Four probe electrical conductivity data of both a room
temperature pressed pellet and crystal of K25Ag45Ce2Te9

as a function of temperature. (B) Thermopower data of a

crystal of K2,5Ag4,5Ce2Te9 as a function of temperature ............. 189

Figure 4.17

Figure 4.18 (A) Four probe electrical conductivity data of a room
temperature pressed pellet of K2_5Ag4,5La2Teo as a
fimction of temperature. (B) Thermopower data of a
room temperature pressed pellet of K25Ag45Ia2Teo as a
firnction of temperature .................................................... 190

Figure 4.19 ORTEP representation of the structure of CuOMEuTez as
seen down the b-axis (70% ellipsoids) ................................. 192

Figure 4.20 ORTEP representation of the structure of KCquuTe4
(70% ellipsoids) viewed down the b-axis .............................. 197

Figure 4.21 ORTEP representation (80% probability ellipsoid) of (A) the
coordination environment around Eu in KCquuTe4, (B) the
coordination environment around K in KCquuTe4, and (C) a
view perpendicular to the Te net in KCquuTe4 ..................... 198

Figure 4.22 (A) Selected area electron diffraction pattern of KCquuTe4
with the electron beam perpendicular to the layers ([001]
direction) showing a twinned 7am x 7bsub domain (i.e.;
two lamb x 7b,“, supercells that are rotated 90° with respect
to one another and superimposed). (B) Selected area electron
diffi‘action pattern of N%,2Ag2,gEuTe4 with the electron
beam perpendicular to the layers ([001] direction) showing
the la x 7h superlattice of single crystal region. (C)
Densitometric intensity scan along the b*-axis of the
electron difli'action pattern of Na0_2Ag2_3EuTe4 (Fig 4.213)
(boxed area in photograph) showing the (-3 k 0) family of
reflections .................................................................. 205

xxiii

 

 

 

'7‘!

‘i~ ‘-
\

 

Figure 4.23

Figure 4.24

Figure 4.25

Figure 4.26

Figure 4.27

Figure 4.28

Figure 4.29

Figure 4.30

Figure 5.1

Inverse molar magnetic susceptibility (l/xM) plotted
against temperature (2-300K) for (A) KCquuTe4 and (B)

Nao,2Ag2_gEuTe4 ........................................................... 208
Diffuse reflectance optical spectra (in the Mid-IR region)
of Nao,2Ag2_gCeTe4 ........................................................ 209

(A) Four probe, electrical conductivity data of room
temperature pressed pellets of KCquuTe4 and NaozAgnguTm
as a function of temperature. (B) Thermopower data of room
temperature pressed pellets of KCquuTe4 and Na0,2Ag2_gEuTe4

as a function of temperature ............................................. 211
ORTEP representation of the structure of K065Ag2Eu1_3 5Te4
(80% ellipsoids) viewed down the a—axis .............................. 214

ORTEP representation of (A) the coordination environment

around Eu in K0.65Ag2Eu.,35Te4 and (B) the coordination
environment around K/Eu in Kn,65Ag2Eu1_35Te4 (90%

ellipsoids for both) ....................................................... 215

ORTEP representation of the Te “net” of KogsAngumsTeii
as seen along the ab plane (80% probability ellipsoids)
highlighting the arrangement of trimers and heptamers ............. 216

(A) Inverse molar magnetic susceptibility (l/xM) plotted

against temperature (2-300K) for K065Ag2EuL35TC4. (B)

Diffuse reflectance optical spectra (in the Mid-IR region)

for K0.65Ag2EU1_35TC4 ...................................................... 223

(A) Four probe electrical conductivity data for single

crystals of Ko,65Ag2Eu.,_-,5Te4 as a function of temperature

(B) Thermopower data for single crystals of K0,65Ag2Eu,,35Te4

as a function of temperature ............................................. 225

ORTEP representation of the structure of K2Ag3CeTe4
viewed down the b-axis (90% probability ellipsoids) ................ 247

xxiv

F
.‘o;\

5.9»
-

 

 

.-
t
V
1
ti
“o
,1-
n‘

 

 

Figure 5.2 (A) Layers of K2Cu2CeS4. (B) Corrugated [AngeTe413‘
layers in KzAg3CeTe4. (C) Inclusion of the third Ag
atoms, between the [AngeTe4]3‘ layers, links them
together into a three-dimensional structure. (D) Tunnel
window projection ........................................................ 248

Figure 5.3 Polyhedra representation of the open channels in
KzAg3CeTe4 with corresponding dimensions ......................... 249

Figure 5.4. ORTEP representation of the structure of szCu3CeTes
as seen down the b-axis (90% ellipsoids) .............................. 253

Figure 5.5 Schematic comparison of the two-dimensional layers of
ZI‘SC3, the one-dimensional :1.;[CeTe5]5 ' chains and the

-‘- [CuzCeTe5]3' chains in szcmcereS ................................ 254
m

Figure 5.6 (A) View perpendicular to the layers of RbZCu3CeTe5,
illustrating how the second Cu atom stitches together

the ~1- [CuzCeTe5]3' chains to form two-dimensional
CD

layers. (B) The distorted [CuTe]', PbO-like layer in
szCU3CCTC§ .............................................................. 255

Figure 5.7 Powder XRD patterns of (A) pristine K2Ag3CeTe4
before ion-exchange (B) Lil + K2Ag3CeTe4, (C)
Nal + KzAggCeTe4, and (D) NH41 + KzAg3C6T64 .................. 261

Figure 5.8 (A) Inverse molar magnetic susceptibility (l/xM)
plotted against temperature (2-300K) for K2Ag3CeTe4.
(B) Diffuse reflectance Optical spectra of KzAg3CeTe4
(in the Mid-IR region) .................................................... 264

Figure 5.9 (A) Inverse molar magnetic susceptibility (l/xM)
plotted against temperature (2-300K) for Rb2Cu3CeTe5.
(B) Diffuse reflectance optical spectra of szCu3CeTe5
(in the Mid-IR region) ................................................... 265

Figure 5.10 (A) Variable temperature, four probe electrical
conductivity data for a single crystal and a pressed pellet
of K2Ag3CeTe4. (B) Variable temperature thermopower
data for single crystals of KzAggCeTe4 ......................................... 267

 

g.

...t.

IVO‘A

..u
-

 

 

 

 

Figure 5.11

Figure 6.1

Figure 6.2

Figure 6.3

Figure 6.4

Figure 6.5

Figure 6.6

(A) Variable temperature, four probe electrical

conductivity data for a single crystal of szCu3CeTe5.

(B) Variable temperature thermopower data for a single

crystal of szCu3CeTe5 ................................................... 268

ORTEP representation of the structure of CUXUTC3

(x = 0.25, 0.33) as seen down the b-axis (80% ellipsoids).
The ellipses with octant shading represent U atoms. The

crossed ellipses represent Cu atoms and the open ellipses

represent Te atoms ........................................................ 280
Extended stuctures of (A) ct-UTe3 and (B) B-UTe3 .................. 283
Extended structure of T10 56UTe3 as seen down the b-axis .......... 284

X-ray powder diffraction patterns of (A) or-UTeg and
(B)-(E) the products of 1U + 3Te heated to 650°C for
2days, 5days, 7days, and 11 days ....................................... 288

Powder x-ray diffraction patterns of (A) elemental copper,
(B) 0.5 Cu + 1.0 a—UTe3 before heating, and (C) 0.5 Cu +
1.0 a—UTe3 after heating ................................................. 292

(A) Selected area electron diffraction pattern of

CuozsUTe; with the electron beam perpendicular to

the layers ([001] direction) showing the incommensurate
superlattice reflections along the a*-axis. (B) Densitometric
intensity scan along the a*-axis of the electron diffraction

pattern (boxed area on photograph) showing the (h10) family

of reflections. The three reflections fiom the sublattice of
Cuo_25UTe3 are indexed. The four weak peaks are from the
superlattice with asuper = 6.25am ........................................ 295

xxvi

'v’.‘~g

s

'I'vh

on!

 

 

Figure 6.7

Figure 6.8

Figure 6.9

Figure 7.1

Figure 7.2

Figure 7.3

Figure 7.4

Figure 7.5

(A) Selected area electron diffi’action pattern of

Cuo_33UTe3 with the electron beam perpendicular to

the layers ([001] direction) showing the incommensurate

superlattice reflections along the a*-axis. (B) Densitometric

intensity scan along the a*-axis of the electron diffraction

pattern (boxed area on photograph) showing the (hkO) family

of reflections. The three reflections from the sublattice of
Cuo_33UTe3 are indexed. The four weak peaks are fiom the
superlattice with a,“per = 6.0a,» .......................................... 297

(A) Variable temperature, four probe electrical conductivity

for bulk crystals of CuxUTe3 (x = 0.25 and 0.33). (B) Variable
temperature thermopower data for bulk crystals of CuxUTe3

(x = 0.25 and 0.33) ......................................................... 301

(A) Variable temperature, four probe electrical conductivity

for a room temperature pressed pellet of a-UTe3. (B) Variable
temperature thermopower data for a room temperature pressed

pellet of a-UTe3 ............................................................ 302

ORTEP representation of the extended structure of
szBaCugTelo as seen down the b-axis (90% ellipsoid
probability) ................................................................ 323

(A) ORTEP representation of the barium filled [Cu3Te12]

cages of szBaCugTelo (50% ellipsoid probability ellipsoid)

and (B) the coordination environment around Rb in

szBaCllgTelo .............................................................. 324

ORTEP representation of the extended structure of . .
CSzBaCllgTCm as seen down the a-axis (90% probabrlity
ellipsoids) ................................................................... 325

Coordination environments around (A) Rb in
szBaCusTew and (B) Cs in CszBaCuriTew ........................... 326

(A) Variable temperature electrical conductivity data

for a single crystal of szBaCugTeto and (B) Variable
temperature thermopower data for a single crystal of 32
RbZBaCIlgTelo ............................................................. 3

xxvii

1‘-
‘s
“:‘t
M. .1
x.‘
\‘0
‘1.
53
5. I'
’1‘”
5. .

’0‘.

has b

 

Figure 7.6

Figure 7.7

Figure 7.8

Figure 7.9

Figure 7.10

Figure 7.11

Figure 7.12

Figure 7.13

Figure 7.14

Figure 7.15

(A) Variable temperature electrical conductivity data

for an ingot of KzBaCugTelo and (B) Variable

temperature thermopower data for an ingot of

KzBaCllgTelo ............................................................... 333

(A) Variable temperature electrical conductivity data and
(B) Variable temperature thermopower data for five ingots
of szBaCugTem (Samples Sl-SS) ...................................... 334

(A) Variable temperature electrical conductivity data and

(B) Variable temperature thermopower data for four ingots
of CszBaCugTelo (Samples S6-S9) .................................... 335

(A) Variable temperature electrical conductivity data

and (B) Variable temperature thermopower data for (a)
szBaCugTelo, (b) szBaCugTelo + 0.lBa, (c) szBaCugTem

+ 0.3Ba, and (d) szBflCUgTelo +0.4Ba ............................... 336

(A) Variable temperature electrical conductivity data

for (a) a pressed pellet of szEUCUgTClo, (b) an ingot

0f szEUCusTew, and (C) an ingot 0f szEUCLQTCIO 'i'

0.2Eu. (B) Variable temperature thermopower data for

(b) an ingot of szEuCugTem, and (c) an ingot of

szEUCU3Tem + 0.2Eu ................................................... 337

Diffuse reflectance optical spectra of (A) szBaCugTem,
(B) CszBaCugTelo and (C) szEUCUgTClo (in the Mid-IR
region) ....................................................................... 339

Heat capacity (J/mol-K) data for (A) four ingots of
szBaCugTem and (A) three ingots of CszBaCugTelo

as a function of temperature ............................................. 342
Heat capacity/T data (J/mol-Kz) vs T2 for (A) four ingots

of szBaCugTelo and (B) three ingots of CszBaCugTem ............ 343
Rama“ SPCCIIa Of CSzBaCllgTClo and szBaCugTelo ............... 345

(A) Inverse molar magnetic susceptibility (llxM)
plotted against temperature (2-300K) for szEuCugTein
and (B) Molar magnetic susceptibility (1M) plotted against

temperature (2-300K) for szBaCugTelo ............................... 343

xxviii

 

Figure 7.16 DTA diagrams of (A) szBaCugTelo, (B) szEUCUgTClo,
and (C) CszBaCugTelo .................................................. 351

DMF
DTA
EDS
ICP

IR
PDF
SAED
SEM
SQUID

TEM

LIST OF ABBREVIATIONS

Dirnethylformarnide

Differential Thermal Analysis

Energy Dispersive Spectroscopy

Inductively Coupled Plasma

Infrared Spectrosc0py

Pair Distribution Function

Selected Area Electron Diffraction

Scanning Electron Microscopy
Superconducting Quantum Interference Device

Transmission Electron Microsc0py

 

 

iitiit

Chapter 1
The Rationale for Combining Coinage Metals and Rare Earth Metals in Molten

Alkali Metal/Polytellu ride Fluxes

A. Introduction

Over the past two decades, we have watched the world undergo a
technological revolution. In doing so, many electronic devices, such as
computers, have become an everyday commodity and very much a necessity to
our lives. Solid state chemistry has certainly played an important role in helping
with this advancement. Such technologies as high density storage batteries,"2
photovoltaics,3 electroluminescence,4 nonlinear optics,5 high Tc superconductors,6
catalysis,7 and thermoelectrics8 depend on the development of new solid state
materials. Therefore, much of the work within the solid state community is
focused either on the improvement of known material for a specific application or
the discovery of new materials for further technological advancements. These
new materials are generally discovered via an “exploratory” approach by
searching for new compounds in previously unexplored areas.

In the past, this “exploratory” approach involved combining high melting
elements together in a vacuum and heating them at very high temperatures. This
is the so-called ceramic method of synthesis. While this proved useful in
discovering new compounds, there were many problems attributed to this method.
First, the reactants never reached a true molten state and therefore the reaction
occurred via diffusion. In order to acquire a homogeneous product, the product
Often had to be reground after the first heating and subsequently reheated. This
“heat -— grind -— heat -— grind” process was continued until the reaction was

complete. Another problem with this method was that, due to the high

temperatures needed for diffusion, only the most thermodynamically stable
products were obtained. The more complex compounds made up of three or four
elements seemed unattainable. Finally, the products formed were often in powder
form, making structure determination difficult if not impossible.

Compared to solution chemistry, where the reactants are able to undergo
“infinite” difl‘usion, solid state chemistry was in dire need of major synthetic
advancements. This is precisely what occurred, leading to such synthetic
techniques as chemical vapor deposition (CVD),9 hydrothermal and solvothermal
synthesis,10 eutetic combination of binary salts,” and molten fluxes. '2 The work
presented in this dissertation is focused mainly on the use of molten fluxes; in

particular, molten alkali metal/polychalcogenide fluxes.

B. Nature of the Polychalcogenide Flux
Only over the past 12 years has the polychalcogenide flux method become
an established technique for discovering new solid state compounds. While
molten salts have been used for over 100 year as a high temperature
recrystallization media for a variety of binary and ternary compounds, '3 it was not
until 1987 when they were used at lower temperatures to synthesize new
compounds.14 Since then, literally hundreds of new compounds have been
reported.15 One advantage to using molten fluxes is that they allow the reaction
System to choose its own route (either kinetic or thermodynamic) without forcing

it to a certain stoichiometry or structure type. Therefore, metastable phases that

 

v -

n'a- ~-

I 1
it -'~
., .

.v -

i

la ‘i'
.J. I
a...

l
‘I'- c
“-....,

could not be synthesized previously are now accessible. This is largely due to the
fact that the flux is a low melting salt and thus the reaction can be carried out at
relatively lower temperatures. The melting points of several alkali
metal/polychalcogenide salts (AzQx) are given in Table 1.2, which illustrates how
the melting points more or less decrease with increasing x value. The flux, once
molten, acts both as a solvent and a reactant, incorporating the chalcogenide
and/or the alkali metal into the final product. In addition, the flux facilitates
crystal growth. Small or poorly formed crystallites can redissolve in the flux and
then reprecipitate as larger, well-formed crystals. This is called the mineralizer
effect. Finally, the crystalline product, albeit powder or crystal form, can easily
be isolated by dissolving the excess flux in simple polar solvents such as methanol

or DMF.

 

 

 

 

 

 

 

 

 

 

 

Table 1.1 Melting points for some known alkali metal/polychalcogenide
(AzQx) species. 16'”
LiZS Lizsz
900.975°c 369°C
Nags N328; Na283 NaZS4 Nazss
1 180°C 490°C 228°C 275°C 252°C
K28 K282 K233 K284 K2S5 K286
840°C 470°C 252°C 145°C 206°C 189°C
RbZS RbZSZ Rb283 RbZS4 Rb285 Rb2S6
530°C 420°C 213°C 160°C 225°C 201°C
Cszsz C3283 C5284 C3285 C8286
460°C 217°C 160°C 210°C 186°C
NaZSe NaZSez Na28e3 Na2$e4 Nazseé
>875°C 495°C 3 13°C 290°C 258°C
K2Se2 K2863 K28e4 K2Se5
460°C 380°C 205°C 190°C
NazTe NazTez NazTe6
953°C 348°C 436°C
KzTe K2T63 K2T64 KzTes KzTes
900°C 429°C 266°C 268°C 264°C
“szre szTe3 szTes
775°C 400°C 270°C
_” CszTe CszTe3 CszTe4 CszTes CszTe6
820°C 395°C 221°C - 237°C 235°C 226°C

 

 

 

 

 

 

 

 

a“

 

'1;
‘-
9? A\ 0
up.» '

 

.1;
t
I
I .
n' "
‘I‘\ ‘
‘1...
u.
. “M
1. .
\ 'a
u ‘5‘
'- l
l' «-
|l M‘
\
2""
‘. .-
W‘-
~‘.
~\“.
-I\.
"b
b
N
'

The reaction between the A2Qx flux and the metals occurs in situ. A typical
reaction mixture consists of AZQ/M/MVQ where the ratio of A2Q/Q is varied fi'om
reaction to reaction. Although it is still unclear the actual mechanism of this
reaction, conceptually it can broken down into two steps. In the first step, the AZQ
species reacts with the Q to form the AZQx flux (see Scheme 1). As a result, long

polychalcogenide chains are formed due to their natural ability to catenate.

Scheme 1
mA2Q+nQ —-(%-> 2111A+ + '[Q—Qx—Ql‘ —N:-/2-I;d—> AwaM’sz

These polychalcogenide chains are made up of two types of chalcogenide atoms: i)
the internal atoms that carry a zero-valent charge and ii) the terminal atoms that
carry each a 1- charge. In the second step, these polychalcogenide chains react
with the metals in the mixture, split, and form a metal chalcogenide fi'amework.
The internal atoms act to oxidize the metal while they themselves are reduced. The
terminal atoms, due to their negative charge, act as Lewis base sites to coordinate
to the metal species.

The length of these polchalcogenide chains and therefore the relative
basicity of the reaction depends strongly on the AZQ/Q ratio. If more Q (or less
A20) is added to the reaction mixture, longer polytelluride chains will form.
Therefore, there are more internal atoms and the reaction mixture will be more

oxidizing, Conversely, if more AzQ (or less Q) is added to the reaction mixture,

'5 0'0 1
k.

.. 5-l
,..>.
i r
‘I u.‘
M
.N" I

t
.1'.
A L-

.J"

2"
\.
-\
k ‘ q
~\.. .
u, I
N
5,
I.
.| I‘
"‘
l
’0‘:
V. I
‘\
.
.
t n
.
n.
'1
t-
l
\t

 

shorter polytelluride chains will form. Therefore, there are more terminal atoms
and the reaction mixture will be more reducing. By varying the AzQ/Q ratio, it is
possible to explore a chemical system under a wide range of conditions. This is

important since some compounds will form only under certain conditions.

C. Synthetic Approach

The system chosen for study in this dissertation was a quaternary one of the
type AwaLnyTez, where M is a coinage metal (Cu, Ag) and Ln is a rare earth
metal (lanthanide or actinide). Explorations in this system with sulfur and
selenium were done previously and the results proved encouraging. 18’” These two
metals were chosen mainly for the fact that they come from very different parts of
the periodic table and thus have very difierent coordination preferences. While
the coinage metals prefer smaller coordination numbers such as 3 or 4, the rare
earth metals desire larger coordination numbers ranging from 6 to 9. Also, the
coinage metals are more covalent in their bonding while the rare earth metals tend
to be more ionic. These differences should maximize the probability that, when
the two metals come together, the compounds formed will possess new structure
types. Indeed, several new structure types were formed (reviewed below) which
show interesting properties such as mixed valency and enhanced conductivity.
Therefore, we decided to extend this chemistry into the telluride system. As will
be illustrated throughout this dissertation, most of the structure types formed are

Strikingly difl‘erent from those found in the sulfide and selenide systems.

D. Review of Quaternary AlM/Ln/Q Phases
The first compounds reported in the A/M/Ln/Q system were KCuCez 86 and
K2Cu2Ce84 by Kanatzidis and Sutorik in 1994.18 Later, the three compounds
KCuLa286, CsCuCegS6, and KCuCeZSe6 were added to the ACuLn2Q6 family. '9
The structure of ACuLn2Q6 is two-dimensional and is composed of [LIISg]
bicapped trigonal prisms that stack in one-dimension by sharing triangular faces to
form chains parallel to the b—axis (see Figure 1.1). Layers are then formed when
neighboring chains share monosulfides. The layers are analogous to the known
phase, ZrSe3,.20 Within the layers, there are tetrahedral sites where the Cu+ ions
reside. Finally, the K“ cations reside in the interlayer gallery. The structure
refinement gave a model in which one copper atom was disordered over two
crystallographically distinct sites. Therefore, it was thought that the Cu+ ions were
statistically distributed over a large excess of tetrahedral sites, leading to possible
ionic conductivity. However, Bensch and coworkers later discovered KCuEuz 86
and found it to possess a 2a x 2b x 2c supercell which removed the disorder from
the original model.” In the superstructure, only one half of the copper sites are
occupied which results in a periodic arrangement of the Cu+ ions along all three
axes.
The two—dimensional structure of K2Cu2CeS4'8 is shown in Figure 1.2. The
anionic layers are composed of [C886] octahedra and [CuS4] tetrahedra. The
[C636] octahedra share edges in one-dimension to form chains. Layers are formed

When these chains alternate with double rows of edge-sharing [CuS4] tetrahedra.

H
O‘.‘;

r W,
.‘.
.

la

n
,~.
,“

‘Q
.
\ J I,
J‘l
lv

 

 

The repeat pattern across the layer is therefore [oct-tet-tet-oct-tet—tet]. The layers
are separated by K+ ions that are stabilized in a seven coordinate monocapped
trigonal prismatic environment of sulfur. This structure type is not particular to
the rare earth metals, as it was found to also exist for NaZCuzlrS...22 Interestingly,
the formal charges on K2Cu2Ce84 cannot be balanced simply by invoking Cu+,
Ce“, and 82'. Three possible formalisms therefore exist: (K+)2(Cu+)2(Ce4*)(Sz')4,
K2(Cu”)2(Ce3+)(SZ')3(S"), or (Kt),(Cu*)(Cu2*)(Ce3*)(sz')4. Magnetic
susceptibility measurements have ruled out the first possibility by verifying the
existence of Ce3+ in the compound. Since Cu2+ is too oxidizing to coexist with 82',
the most logical formalism was chosen to be the second one. This mixed 3278"
model places holes in the sulfur p-band, which predicts high conductivity. This
was not experimentally observed, however, and it was speculated from this that
the narrow valence bands in the compound are acting to limit the mobility of the
8" holes and small polarons could be forming which act to frustrate the carriers.
Many other compounds found in the A/M/Ln/Q system were found to have
the general formula, AMLnQ3, yet they do not all possess the same structure type.
There are four known structure types that exist for this stoichiometry, see Figure
1.3. Of these four, however, structure type A is the most stable. In the past five
years, many members have been added to this large family, including KCuUSe3,l9
CsCuCeS3,l9 CsCuUTe3,23 BaCuLnS3 (Ln = Sc, Y, Gd, 130,24 BaCuLnSe; (Ln = Y,
EFL“ BaCuDyTe3,25 BaCuLnTe; (Ln ;= Y, La, Pf, Nd» Yb)” BaAgNdS3,24

BaAgLnSe3 (Y, La, Er),24 BaAgLnTe; (Ln = Y. La. Gd).26 and BaAquSe326.

.-.. ‘.

LAC

nu".

’AU .k

fl‘t
_‘ .n
. .
| .
5‘", ,-
.~_’g_ ‘

.
WI- :-

an»

‘N
.y
‘.b
'l
n -
‘t
7'
.
1':
7
.;‘_‘ a
‘7‘. r
'.
~
I
'
u
\'I
y I"
e
X
1
~-
'1.

 

 

Much like K2Cu2CeS4, this AMLnQ; structure type [A] is comprised of [Lan]
octahedra and [MQ4] tetrahedra. However, the [LnQé] octahedra now edge-share
with “single” rows of [MQ4] tetrahedra instead of the “double” rows found in
KZCuZCeS4. Therefore, the repeat pattern across the layer is [oct-tet-oct-tet]. A
slightly distorted version of this structure type also exists in which some of the
bonds are lengthened, due to the larger rare-earth elements used. This causes the
symmetry to drop for the above phases from Cmcm to Puma. The compounds that
crystallize in this space group are BaCuLnS3 (Ln = Ce, Nd),2427 [i-BaCuLaSe3;,2"’27
BnCuCeSe3,24 and BaCuLaTe326.

The second structure type [B] having the general formula AMLnQ; is
shown in Figure 1.3B. The structure is related to the previous two structure types
in that the layers are comprised of [LnQé] octahedra and [MQ4] tetrahedra. In this
case, however, the layers are made up of alternating pairs of octahedra and pairs
of tetrahedra. Therefore, the repeat pattern across the layer is [oct-oct-tet-tet-oct-
oct]. Interestingly, the compounds that adopt this structure type are more
specifically of the type AMMQ3, since both metals are transition metals (e.g.,
NaCuTiS3, NaCquSe3, NaCquTe3).22 However, it is not unreasonable to assume
that a rare earth metal could take the place of the octahedral transition metal.

The third structure type [C] having the formula AMLnQ3 is shown in
Figure 1.30 It is three-dimensional and, interestingly, is not related to either of
the above structure types [A or B]. To date, only one compound is known to adopt

this structure type, BaAgErS3.‘5“’28 It crystallizes in the monoclinic space group,

10

W' I

l

\v d. W
.5 l

 

\. . .
o, 1

\
k..- r

"-

-q. s

.o,

1.x

 

 

 

 

 

 

 

C2/m and is made up of [EfSé] octahedra and [Ang trigonal bipyramids. The
[131.86] octahedra share edges in a zigzag manner to form double-chains that run
down the b—axis. These chains then share comers along the a-axis to form two-
dimensional layers. The layers are firrther connected into a three-dimensional
framework through pairs of corner sharing [AgSs] trigonal bipyramids (Ag289
units). The Bay ions occupy the sites inside the channels and are stabilized in a
7-coordinate monocapped trigonal prsmatic environment.

The fourth structure type [D] having the formula AMLnQ3 is shown in
Figure 1.3D. The two compounds that adopt this structure type (BaCuLa83 and
(ii-BaCuLaSe3)2“7 crystallize in the orthorhombic space group, Puma. However,
since this structure type is different from that of the second structure type [B],
which also crystallizes as ana, this structure will be denoted as ana (11) while
structure type [B] will be denoted as ana (I). The structure is made up of [LaS7]
monocapped trigonal prisms that make edge sharing connections with [CuS4]
tetrahedra to form a three-dimensional fiamework. Interestingly, a-BaLaCuSe3
can be transformed to the B-phase of BaLaCuSe; (structure type A) by annealing
at elevated temperatures. Conversely, the (at—phase can be generated from the
5*phase by mechanical grinding. The reason that these two structures can be
converted back and forth is that they are structurally very similar. In the (it—phase
the La atoms are Geocrdinate octahedral while in the B—phase, they are 7 -

coordinate monocapped trigonal prismatic. The transition from the a—phase to the

11

 

 

 

 

B—phase involves a distortion of the octahedral La atoms so that they may bond to
another sulfur atom from adjacent layers.

Another structure type that was found in the A/M/Ln/Q system was
KCquzsa.29 The structure of this compound is shown in Figure 1.4 and is again
made up of [Gde octahedra and [CuS4] tetrahedra. Structurally, it is very similar
to BaAgEr83 (structure type [C] of AMLnQ3). In BaAgErS3, double chains of
[[5de octahedra are connected into a three-dimensional framework through pairs
of corner sharing [AgSs] trigonal bipyramids (Ag289 units). These same [LnQ6]
double chains exist in KCquzsa, only now they are connected into a three-
dirnensional framework by single [CuS4] tetrahedra units. As a result, there is less
Cu (per Gd) in the chemical formula and the channels that run through this
structure are smaller. Although the channels are smaller in KCqu284, the
coordination environment inside the channel is larger. While the Ba2+ ions in
BaAgErS; are monocapped trigonal prismatic, the K+ ions in KCqu284 are
bicapped trigonal prismatic. This could be attributed to the fact that the ionic radii
of 1C (1.52 A) is slightly larger than that of Ba2+ (1.49 A). However, a more
logical explanation may be the different coordination environments around the
coinage metals. In BaAgErS3, the silver atoms are trigonal bipyramidal while the
COpper atoms in KCqu2$4 are tetrahedral. Therefore, the different sized tunnels
that host the alkali/alkali earth metals could simply be a manifestation of the

different coordination environments around the coinage metals.

12

 

 

 

 

Finally, the compound K6Cu12U2815 was synthesized in our lab by AC
Sutorik and further characterized by myself.30 This three—dimensional compound
is shown in Figure 1.5. Although the structure is too complex to discuss in much
detail here, the basic building block is one of a [1186] octahedra which edge shares
with six [CuS3] trigonal planar units. Much like KZCuZCesa, the charges cannot be
balanced on this compound without invoking some 8278" mixed valency. This
mixed valency again places holes in the valence band and predicts high
conductivity. Unlike KzCu2CeS4, however, where the narrow valence bands limit
the carrier’s mobility, K6Cu12Ule5 shows metallic behavior. This can be

attributed to its highly covalent, three-dimensional framework.

13

 

(I? 3

c ‘> 3

 

 

 

Cu

 

 

 

 

 

 

Figure 1.1 Extended structure of ACuLn2Q6 as seen down the b-axis. Black

Clrcles 1' ePresent Ln atoms, striped circles represent Cu atoms, and large open

“Mes represent A and Q atoms.

14

a” U

15‘

 

 

 

 

 

F0
lgure 1.2 Extended structure of K2Cu2CeS4 as seen down the b-axis. Black

circl . .
es rePresent Ce atoms, striped crrcles represent Cu atoms, and large open

Clrcles r“Bimisent K and S atoms.

15

 

Figure 1'3 (A) Cmcm structure type of AMLnQ3 (e.g.; KCUUSC3) and (B)
Puma (1) Structure type of AMLnQ; (e.g.; NaCuTiS3). The small open circles
represent M atoms, the black circles represent Ln atoms, and the large open circles
rcpresem A and Q atoms.

16

 

Figure1,3 continued (C) C2/m structure type of AMLnQ3 (8-8-3
B

aAgErSs) and (D) ana (11) structure type of AMLnQ3 (ea; 3810114353)-
The small open circles represent M atoms, the black circles represent Ln

ato .
m3, and the large open crrcles represent A and Q atoms.

17

Abra. u

ANN come

-NVANAr v

WWWVAAtv

I IRIIlL-

l.

 

Black

Figure 1.4 Extended structure of KCquZS4 as seen down the a-axis.
circles represent Gd atoms, striped circles represent Cu atoms, and large open

circles represent K and S atoms.

18

rt)

 

 

 

 

 

 

Figure 1.5 ORTEP representation of the extended structure of K6Cul2U2815.
Crossed ellipses represent K atoms, large open ellipses represent Cu atoms, small
Open ellipses represent U atoms, and octant shaded ellipses represent U atoms.

19

 

E. Te net distortions

In moving from Q = S, Se to Q = Te, it is important to understand the
differences between these chalcogens. One important difference is the greater
tendency for the latter to associate through Te -— Te bonding interactions because
of the more diffuse nature of its orbitals. Tellurium is less electronegative and can
therefore stabilize longer than normal bond distances. Roald Hoffrnann recently
described tellurium’s behavior as a “constant flirtation with other tellurium
partners, in the range between a bond and no bond”.3 1 This has been illustrated in
the alkali metal rich tellurides32 where such Te — Te “flirting” has resulted in the
formation of one-dimensional chains (infmite33 and spirocyclic“), cyclohexane-
like Te6 rings,” and puckered crown shaped Teg rings”. When a transition metal
or a rare earth metal is added to the synthesis, two-dimensional Te nets have been
observed, for example in NdTe337 and K0,33Bao.67AgTe23°. This is not to say that
the units found in the alkali metal rich tellurides are not found here. In fact, such
compounds as (lt-—UTe339 and AThzTeG (A = Cs, Cu)40 possess one-dimensional
infinite chains in their structure. However, these Te nets are of particular interest
because they have been found to undergo structural distortions that not only result
in interesting superstructures but also have a drastic affect on the physical
pr0perties of the material. This is illustrated in the binary rare earth telluride
compounds, LITTC3. The extended structure of LnTe3 as viewed down the c-axis is

shown below.“

20

 

 

 

 

at
L if

The two-dimensional structure is made up of corrugated, cubic rare earth telluride

 

 

 

 

slabs that alternate with planar Te nets. The Ln atoms (black circles) are each
coordinated to nine Te atoms (white circles) in a monocapped square antiprismatic
geometry. The Te net appears as perfectly square with all Te-Te bond distances
equal around 3.1A. This Te net, shown below, is what will be referred to as an

“ideal” configuration.

 

“Ideal” Te net

In terms of physical properties, if the Te net is truly “id ”, the material should be
metallic. This is because the electronic bands that give rise to a material’s
conductive nature are entirely derived from this square Te net. In other words, the

energy levels at the Fermi Level are primarily made up of Te p—orbitals and the

21

in... u.
- r
t4 ‘_
il‘ J.
' «an.
o

:-.~«,

“in ..

htmt

lv'
5

bands associated with the rare-earth telluride slab do not contribute at all to the
Fermi surface. Due to the more diffuse nature of the Te p-orbitals, there is good
overlap and the electron density is considered to be “delocalized” across the net.
This gives rise to the high conductivity. However, since the average charge per Te
atom in these nets is usually less than 2' (0.5' in the case of LIITC3), the nets are
considered to be “electron deficent” and are therefore susceptible to distort.
Below is a cartoon illustration of the electronic band structure of a material with a

Te net in an “ideal” configuration and one that has a “distorted” Te net.

 

 

 

 

 

 

 

 

 

Scheme 1

“Ideal” Te net “Distorted” Te net
. ‘ CB

67) \ g

EF 3 Ete— EF
m \_ Lu
s -———’
d'st rt
1 O ‘ VB
.11;— it

o k 2am 0 k ”Super
Metal Semiconductor

This distortion in the Te net lowers the total energy of the system by decreasing

the energy of the filled Te p~orbitals which localizes the electron density into fully

occupied bonding orbitals. Consequently, a gap is opened up at the Fermi Level

and the physical properties of the material changes from metallic to

semiconducting.

22

 

For LnTe; (Ln = Nd, Sm), the standard crystallographic determination
presented an “ideal” Te net.37 Consistent with this model, tight binding
calculations predicted metallic properties. However, the conductivity
measurements reported for pressed pellets of LaTe3 and ErTe3 suggest
semiconducting behavior.42 This led Lee and DiMasi to re-examine these LnTe;
materials to try and better correlate the structure with the properties.43 By using
electron diffraction, they identified superlattice reflections indicating the presence
of incommensurate distortions, consistent with modulations in the square Te nets
in LnTe3 (Ln = La, Sm, Gd, Tb, Dy, Ho, Er, Tm).44

Certainly, these distortions are not specific to square nets. However, they
are most commonly found to exist in low-dimensional compounds. The low
dimensionality simply makes the compounds more susceptible to distort.
Examples include the chainlike transition metal trichalcogenides,“ the layered
transition metal dichalcogenides,46 and the red47 and purple48 bronzes. More
recent examples include K3CU3S6,49 V3Te4,50 and SmTemfl LaSe,_o,52’53
DySe1_34,53’54’55 and RbDy3Se352’53’54. Distorted chalcogen nets have been observed
for both stoichiometr‘ic56 [Lan] and chalcogen deficients7 [LnQ2_x] compounds.
The chalcogen deficiency leads to vacanciess8 in the net, which drives the system
to distort.

As far as the chalcogen nets are concerned, there are two main ways for
them to distort, via a Charge Density Wave (CDW) or a Site Occupancy Wave

(30W). A CDW type distortion is defined as a sinusodial atomic displacement,

23

 

3"? .-
’n~.. L

"I

5.::

L‘udh

4P"
"6“”

'1‘ .-

._~ 7“

.,‘

combined with an electron-phonon coupling, which together produce either a gap
or a deep valley at the Fermi level by lowering the energy of the occupied states
while at the same time raising the energy of the unoccupied states.59 The wave
vector of the distortion, or modulation, can occur in any direction in the plane of
the Te net. Another way to think about a CDW type distortion is as a Jahn-Teller
type distortion in which large displacments of atoms cause the atomic coordination
to be reduced.60 This is evident if we think about a simple system like elemental
tellurium. Te has two unoccupied states in degenerate p—orbitals and requires two
nearest neighbors in order to fill its valence shell through covalent bonding.
Therefore, the desired coordination number for Te is two. There are two phases of
elemental tellurium?” a high pressure phase and an ambient pressure phase.
Under high pressure, the coordination number of Te is six (octahedral). However,
at ambient pressure, the coordination number drops to two in the form of rings or
chains. This is an indication that there in fact is a significant driving force for
these atoms to reduce their coordination.

An example of a CDW type distortion is shown below, see Scheme 2.

 

24

Depicted on the left is a Te net in an “ideal” configuration with a unit cell that
correctly describes its periodicity. If the atoms above the arrows distort in the
direction described, a new Te net results which is shown on the right. It is made
up of alternating infinite Te chains and monotellurides and the coordination within
the net drops from four to two and zero. Consequently, a new, larger unit cell
(may be commensurate or incommensurate) is needed to redescribe the periodicity
of the Te net, which is referred to as the supercell. The original unit cell is called
the subcell.

Another way that these Te nets can respond to various electronic situations
is via a Site Occupancy Wave (SOW). In this case, some of the Te atoms are

actually removed from the solid state lattice, creating ordered vacancies. An

example of this is shown below in Scheme 3.

 

 

 

Again, on the left is a Te net in an “ideal” configuration. If the Te atoms that are
highlighed are removed from the net, the resulting net is shown on the right. This

time, '
only some of the Te atoms experience a lowering of their coordination

25

. v‘di
,3...»

w. ‘-.>

...'
..et\..

‘a " .—

58 h.\

i..“""

Nehru

I . .l
l'

- u. ..

I
-‘

 

\ .
. I
v
|
,,. h
l.
' HI
1
v

number. As in the case of a CDW type distortion, a larger unit cell is needed to
redescribe the periodicity of the Te net, which is the supercell.
In either case (CDW or SOW), a superstructure exists that better describes
the true picture of the compound. Since the nature of the Te net dictates the
electrical properties of the material, it is very important to achieve a structural
model that is as close to the truth as possible. Otherwise, wrong correlations could
be made between the structure and the prOperties and even worse, wrong
conclusions could be made about the chemistry. As will be shown throughout this
dissertation, many of the compounds found in the A/M/Ln/T e system possess Te
nets that undergo CDW or SOW type distortions. In fact, what we have learned in
doing this research is that perfectly square Te nets are quite unstable.
Unfortunately, however, the crystallographic reflections that give rise to the
supercells that describe these distorted Te nets are oftentimes so weak that they
cannot be detected by standard X-ray diffraction techniques. In these cases,
electron diffraction methods have been employed to try and elucidate the existence
and identity of the supercells. In addition, conductivity measurements have been
invaluable in helping to shed light on the situation. One must be careful, though,
when interpreting the conductivity measurements alone. If, for example, a
material is determined to be metallic, one might conclude that there is no
distortion in the Te net. Yet, the more subtle distortions only partially open a gap

at the Fermi level, causing the material to remain metallic. Therefore, it is best to

26

couple the conductivity measurements with electron diffraction before making any

conclusions.

27

hire

 

 

 

 

 

 

References

 

(a) The Sodium-Sulfia' Battery, Sudworth, J .L.; Tilley, A.R. Eds., Chapman
and Hall, New York, 1985. (b) Fischer, W. Mater. Res. Soc. Symp. Proc.
1989, 135, 541. (c) Powers, R.W.; Karas, B.R. J. Electrochem. Soc. 1989,
136, 2787.

(a) Whittingham, MS. in Solid State Ionic Devices, July 18-23, 1988,
Singapore; Chowdari, B.V.R., Radhakrishna, S., Eds.; World Scientific:
Singapore, 1988; pp 55-74. (b) Bowden, W.L.; Bamette, L.H.; Demuth,
D.L. J. Electrochem. Soc. 1989, 136, 1614. (c) Murphy, D.W.; Trumbore,
RA. J Electrochem. Soc. 1976, 123, 960.

(a) Etrnan, M.; Katty, A.; Levy-Clement, C.; Lemasson, P. Mater. Res. Bull
1982, 17, 579. (b) Katty, A.; Soled, S.; Wold, A. Mater. Res. Bull. 1977,
12, 663.

Ballrnan, A.A.; Byer, R.L.; Eimerel, D.; Feigelson, R.S.; Feldman, B.J.;
Goldberg, L.S.; Menyuk, N.; Tang, C.L. Applied Optics 1987,
26, 224.

(a) West, A.R. Solid State Chemistry and Its Applications; John Wiley &
Sons: New York, 1984, p 540-552. (b) Xia, Y.; Chen, C.; Tang, D.; Wu, B.
Adv. Mater. 1995, 7, 79.

(a) Malik, S.K.; Vijayaraghavan, R. Rev. Solid State Sci. 1988, 2, 75. (b)
Chevrel, R.; Sergent, M.; Prigent, J. J. Solid State Chem. 1971, 3, 515. (c)
Hebard, A.F.; Rosseinsky, M.J.; Haddon, KC; Murphy, D.W.; Glarum,
S.H.; Palstra, T.T.M.; Ramirez, A.P.; Kortan, A.R. Nature 1991, 350, 600.

(a) Chianelli, R.R.; Pecoraro, T.A.; Halbert,T.R.; Pan, W.-H.; Stiefel, E.I. J.
Catal, 1984, 86, 226. (b) Pecoraro, T.A.; Chianelli, RR. J Catal. 1981,
67, 430. (c) Harris, 8.; Chianelli, R.R. J. Catal. 1984, 86, 400.

Rowe, D.M.; Bhandari, C.M. Modern Thermoelectrics, Holt, Rinehart, and
Winston, London, 1983, p. 103. (b) Borkowski, K; Pyzluski, J. J. Mater.
Res. Bull. 1987, 22, 381.

Steigerwald in Inorganometallic Chemistry, T.P. Fehlner, Ed Plenum
Press, New York, 1992, 333.

Rabenau, A. Angew. Chem. Int. Ed. Engl, 1985, 24, 1026.

28

.l\l. «(\J flu “Iv

H4

 

ll

l7

I8

20

21

22

(a) Chen, L.R. Commun. Theor. Phys. 1992, 17 (3), 285. (b) Chen, L.R.
Commun. Iheor. Phys. 1992, 13 (2), 147.

(a) Elwell, D.; Scheel, H.J. Crystal Growth fiom High-Temperature
Solutions; Academic Press: London, 1975. (b) Molten Salts Handbook;
Janz, G.J., Eds.; Academic Press: New York, 1967. (c) Mamantov, G.
Molten Salts; Marcel Dekker: New York, 1969.

(a) Scheel, H.J. J Cryst. Growth 1974, 24/25, 669. (b) Sanjines, R.; Berger,
H.; Levy, F. J Mater. Res. Bull. 1988, 23, 549. (c) Garner, R.W.; White,
W.B. J Cryst. Growth 1970, 7, 343.

Sunshine, S.A.; Kang, D.; Ibers, J .A. J. Am. Chem. Soc., 1987, 109, 6202.

(a) Kanatzidis, M.G. Chem. Mater. 1990, 2, 353. (b) Sutorik, A.C.;
Kanatzidis, M.G. Prog. Inorg. Chem. 1995, 43, 151. (c) Wu, P.; Ibers, J.A.
J. Alloys Camp. 1995, 229, 206. (d) Pell, M.A.; Ibers, J .A. Chem.
Ber./Recueil 1997, 130, 1. (e) Narducci, A.A.; Ibers, J.A. Chem. Mater.
1998, 10, 2811.

(a) Pearson, T.G.; Robinson, P.L. J. Chem. Soc., 1931, 1304. (b)
Mathewson, G.H. J. Am. Chem. Soc., 1907, 29, 867. (c) Klemm, W.;
Sodomann, H.; Langmesser, P. Z. Anorg. Allg. Chem, 1939, 241, 281. (d)
Gmelin ’s Handbuch der Anorganischen Chemie, Verlag Chemie,
Weiheim/Brgstr, F RG 1966, Sodium, Suppl. Part 3, p. 1202-1205 and
references therein.

Binary Alloy Phase Diagrams (Ed: T.B. Massalski), William W. Scott, Jr.,
Materials Park, Ohio 1990, pp 1395, 2392, 2737, and 3193.

Sutorik, A.C.; Albritton-Thomas, 1.; Kannewurf, C.R.; Kanatzidis, M.G. J.
Am. Chem. Soc. 1994, 116, 7706.

Sutorik, A.C.; Albritton-Thomas; Hogan, T.; Kannewurf, C.R.; Kanatzidis,
M.G. Chem. Mater. 1996, 8, 751.

(a) Kronert, W.; Plieth, K. Z. Allg. Anorg. Chem. 1965, 3, 207. (b)
Furuseth, S.; Brattas, L.; Kjekshus, A. Acta. Chem. Scand 1975, A29, 623.

Bensch, W.; Dilrichen, P. Chem Ber. 1996, 129, 1489.

Mansuetto,M.F.;1bers, J.A. J. Solid State Chem. 1995, 130, 1.

29

 

23

24

25

26

27

28

29

30

3l

32

33

34

35

36

37

38

Cody, J.A.; Ibers, J.A. Inorg. Chem. 1995, 34, 3165.

Wu, P.; Christuk, A.E.; Ibers, J.A. .1 Solid State Chem. 1994, 110, 337.
Huang, F.-Q.; Choc, W.; Lee, S.; Chu, J.S. Chem. Mater. 1998, 10, 1320.
Yang, Y.; Ibers, J.A. J. Solid State Chem. 1999, 147, 366.
Christuk, A.E.; Wu, P.; Ibers, J .A. J. Solid State Chem. 1994, 110, 330.
Wu, P.; Ibers, J.A. J. Solid State Chem. 1994, 110, 156.

Stoll, P.; Dfirichen, P; Neither, C.; Bensch, W. Z. Anorg. Allg. Chem. 1998,
624, 1807.

Sutorik, A.C.; Patschke, R.; Schindler, J .; Kannewurf, C.R.; Kanatzidis,
M.G Chem. Eur. J., In press (2000).

Hoflinan, R. American Scientist 1996, 84, 327.

(a) Sheldrick, W.S.; Wachhold, M.; Jobic, S.; Brec, R.; Canadell, E.
Advanced Materials 1997, 9, No 8, 669. (b) Kanatzidis, M.G. Angew.
Chem. Int. Ed Engl. 1995, 34, No. 19, 2109

LizTea: Bottcher, P.; Keller, R. J. Less Common Met. 1985, 109, 311 and
references therein.

szTesz Bottcher, P.; Kretschmann, U. J. Less Common Met. 1983, 95, 81.
RbTe6: Sheldrick, W.S.; Schaaf, B. Z Natwforsch. B 1994, 49, 993.

ngTezzz Sheldrick, W.S.; Wachhold, M. Angew. Chem. 1995, 107, 490;
Angew. Chem. Int. Ed Engl. 1995, 34, 450.

(a) Norling, B.K.; Steinfink, H. Inorg. Chem. 1966, 5, 1488. (b) Noel, H.;
Levet, J .C. J. Solid State Chem. 1989, 79, 28.

(a) Zhang, Z.; Li, J .; Foran, B.; Lee, S.; Guo, H.-Y.; Hogan,T.; Kannewurf,
C.R.; Kanatzidis, M.G. J. Am Chem. Soc. 1995, 117, 10513. (b) Hanko,
J.A.; Kanatzidis, M.G.; Evain, M.; Gourdon, 0.; Boucher, F.; Petricek, V.
Inorg Chem, In press.

30

 

 

39

40

41

42

43

45

46

47

48

49

Breeze, E.W.; Brett, N.H.; White, J. J. Nucl. Mat. 1971, 39, 157.

(3) Cody, J.A.; Ibers, J.A. Inorg. Chem 1996, 35, 3836. (b) Narducci, A.A.;
Ibers, J.A. Inorg. Chem. 1998, 37, 3798.

Stowe, K.Z. Anorg. Allg. Chem. 1996, 622, 1419.

(a) Ramsey, T.H.; Steinfink, H.; Weiss, E.J. J. Appl. Phys. 1965, 36, 548.
(b) Haase, D.J.; Steinfink, H. .1 App]. Phys. 1966, 37, 2246.

DiMasi, E.; F oran, B.; Aronson, M.C.; Lee, S. Chem. Mater. 1994, 6, 1867.

(a) DiMasi, E.; Aronson, M.C.; Mansfield, J.F.; Foran, B.; Lee, S. Phys.
Rev. B. 1995, 55(20), 14516. (b) Gweon, G.H.; Denlinger, J.D.; Clack, J .A.;
Allen, J.W.; Olson, C.G.; DiMasi, E.; Aronson, M.C.; Foran, B.; Lee, S.
Phys. Rev. Lett. 1998, 81, No. 4., 886

(a) NbSe3: Comes, R.; Lambert, M.; Launois, H.; Zeller, H.R. Phys. Rev.
B. 1973, 8, 571. (b) ZrTe3: Eaglesham, D.J.; Steeds, J.W.; Wilson, J.A. J.
Phys. C..' Solid State Phys. 1984, 17, L697.

(a) Peierls, R.E.; Quantum Theory of Solids, Oxford University Press,
Oxford, England, 1955. (b) Frohlich, H. Proc. R. Soc. A. 1954, 223, 296.
(c) Electronic Properties of Inorganic Quasi-One-Dimensional
Compounds, Eds. P. Monceau 1985. (d) Whangbo, M.—H.; Canadell, E. J.
Am. Chem. Soc. 1992, 114, 9587. (e) Rovira, C.; Whangbo, M.-H. Inorg.
Chem. 1993, 32, 4094. (f) Whangbo, M.-H.; Ren, J.; Canadell, E.; Louder,
D.; Parkinson, B.A.; Bengel, H.; Maganov, S.N. J. Am. Chem. Soc. 1993,
115, 3760. (g) Liang, X.; Lieber, C.M. Science 1989, Vol 243, 1703 (h)
Burk, B.; Thompson, R.E.; Clarke, J .; Zettl, A. Science 1992, Vol 257, 362.

(a) A033MoO3: Hillenius, S.J.; Coleman, R.V.; Fleming, R.M.; Cava, RJ.
Phys. Rev. B. 1981, 23, 1567. (b) BaOJSWO33 Canadell, Es Whangbo, M.-
H.1norg. Chem. 1994, 33(9), 1864.

(a) KM06017, N309MO6017, Lio_9MO6O|7I Dumas, J., Schlenker, C. Int. J.
Mat. Phys. 1993, 7, 4045. (b) Whangbo, M.-H.; Canadell, Es Foury. P.;
Pouget, J.-P. Science 1991, Vol 252, 96.

Sato, H.; Kojima, N.; Suzuki, K.; Enoki, T. J. Phys. Soc. Japan 1993, 62,
647.

31

 

'\ f VII] 111‘ h .l k

 

50

51

52

53

55

56

57

S8

59

61

Ohta, S.; Kaneko, T.; Yoshida, H. J. Phys. Soc. Japan 1990, 59, 3827.
Park, S.-M.; Park, S.J.; Kim, S.-J. J. Solid State Chem. 1998, 146, 300.
Lee, S.; Foran, B. J Am. Chem. Soc. 1994, 116, 154
Lee, S.; Foran, B. J Am. Chem. Soc. 1996, 118, 9139.
Foran, B.; Lee, S.; Aronson, M.C. Chem. Mater. 1993, 5, 974.

van der Lee, A.; Hoistad, L.M.; Evain, M.; Foran, B.J.; Lee, S. Chem.
Mater. 1997, 9, 218.

(a) Maroon, J.-P.; Pascard, R. CR. Acad. Sc. Paris 1968, 266, 270. (b)
Bénazeth, S.; Carré, D.; Laruelle, P. Acta. Cryst. 1982, B38, 33. (c) Chen,
J.H.; Dorhout, PK. .1 Solid State Chem. 1995, 117, 318.

(a) Lin, W.; Steinfmk, H.; Weiss, E.J. Inorg. Chem 1965, 4, 877. (b)
Ramsey, T.H.; Steinfink, H.; Weiss, E.J. Inorg. Chem. 1965, 4, 1154. (c)
Wang, R.; Steinfink, H.; Bradley, W.F. Inorg. Chem. 1966, 5, 142. (d)
Webb, A.W.; Hall, H.T. Inorg. Chem. 1970, 9, 1084. (e) Cannon, J .F .; Hall,
H.T. Inorg. Chem. 1970, 9, 1639. (f) DiMasi, E.; Foran, B.; Aronson, M.C.;

Lee, S. Phys. Rev. B. 1996, 54(19), 13587.

(a) Plambeck-Fischer, P.; Abriel, W.; Urland, W. J. Solid State Chem.
1989, 78, 164. (b) Urland, W.; Plambeck-Fischer, P.; Grupe,M. Z
Naturforsch 1989, 44b, 261. (c) Grupe, M.; Urland, W. J. Less-Common
Met. 1991, 170, 271. ((1) Kim, S.-J.; Oh, H.-J. Bull. Korean Chem. Soc.

1995, 16, 515.
Wilson, J.A.; DiSalvo, F.J.; Mahajan, S. Adv. Phys. 1975, 24, 117.

(a) Albright, T.A.; Burdett, J .K.; Whangbo, M.-H. Orbital Interactions in
Chemistry, Wiley, New York 1985. (b) Whangbo, M.-H.; Canadell, E. J.

Am. Chem. Soc. 1992,114, 9587.
Martin, R.M.; Lucovsky, G.; Helliwell, K. Phys. Rev. B. 1976, 13, 1383.

32

 

Rm

 

Chapter 2

Reactions of Rare Earth Metals in Molten Alkali Metal/Polytelluride Fluxes:

Discovery of the ALn3Te8 Family (A = Cs, Rb, K; Ln = Ce, Nd)

33

A. Introduction
Many high symmetry, low-dimensional compounds contain stacking layers
which can be described as square lattice networks composed of one element.

Examples include compounds of the Cme structure type,1 the related ZrSiSe
structures,2 the PhD and anti-PbO structures,3 the BaAl4,4 ThCr28i25 and
CaBezGez“ structure types, the SanSb2 structure,7 and other less popular structure

types.8 The chemical, physical, and electronic properties of these compounds are
largely decided by these square nets, and by their interaction with the remaining
part of the structure. However, only a few were known for tellurium at the start of
this research, e.g., LnTez, anTes, and LnTe3,9 Cs'l'hzTeé,lo K0.33Ba0.67AgTe2,ll and
Cs3Te22‘2. These square nets can have different electronic structures, which can

'3 These distortions

lead to instabilities and structural distortions within the nets.
are associated with several interesting physical phenomena such as charge density
waves and can lead to anomalies in the charge transport properties. When the
formal oxidation state of all Te atoms in the net is -2, a stable square net is
observed (e.g. NaCuTe).14 In this case, the term “square net” is used to describe
the arrangement of the Tez' ions in which there is no bonding at all between the

2. . . . .
Te ions. However, when the formal oxrdation state is less than -2, or when there

are atomic vacancies in the square net, structural distortions are possible leading to
. n- . a .
Te°°Te bonding interactions and the formation of Tex spec1es. These distortions

are manifested through the formation of a superstructure with respect to the ideal

34

u
.
3......
t

i"
ll
r_

l
A.“

- r
M‘-

 

 

square net.ll A family of compounds having the formula ALn3Te8 (A = Cs, Rb, K;
Ln = Ce; Nd) has been discovered which displays a defect square Te net and an

unprecedented intense charge density wave, leading to the formation of infinite

2. -
zig-zag (T c2 )n chains and Te32 anions. Interestingly, this charge density wave
has recently been predicted on theoretical grounds, ”’16 and this report constitutes

the first experimental confirmation.

B. Experimental Section
1. Reagents — The following reagents were used as obtained: Potassium
metal, analytical reagent, Spectrum Chemical Mfg. Corp., Gardena, CA;
Rubidium metal, 99.5%, Alfa Aesar, Ward Hill, MA.; Cesium metal, 99.98%,
Alfa Aesar, Ward Hill, MA; Copper metal, electrolytic dust, Fisher Scientific,
Fairlawn, NJ; Cerium metal, < 250 mesh, Alfa Aesar, Ward Hill, MA; Neodynium
metal, < 250 mesh, Alfa Aesar, Ward Hill, MA; Tellurium powder, 100 mesh,
99.95% purity, Aldrich Chemical Co., Milwaukee, WI; N, N, -
Dimethylformamide (DMF) was used as obtained in analytical reagent grade from
Aldrich Chemical Co., 99.8% purity, Milwaukee, WI-
Potassium T elluride, K 2T e — The following procedure was modified from
that given in the literature." 11.50g (0.29 mol) K was sliced in an N2 filled
glovebox and combined with 18.50g (0.14 mol) Te in a 1000 mL single neck

round bottom flask. This mixture represents a slight excess of K and slight

35

 

deficiency of Te. The flask was connected to a glass adapter with a stopcock joint
and removed from the glovebox. The flask and adapter was then connected to a
condenser apparatus and chilled to -78°C using a dry ice/acetone bath.
Approximately 800mL of NH3 were condensed, under an N; atmosphere, onto the
reagents, giving a purple solution. The solution was stirred via a Teflon coated
magnetic stir bar and the reaction mixture was maintained at -78° for up to 24
hours. The dry ice was then removed and the NH3 was allowed to evaporate off as
the flask warmed up to room temperature under a constant flow of N2
(approximately 10 hours). A second portion of NH3 was added and the process
was repeated to ensure complete reaction of the reagents. The resulting pale
yellowish-grey powder was evacuated on a Schlenck line for approximately 5
hours and taken into an N2 filled glovebox where it was ground to a fine powder.
Due to its propensity to decompose even under an inert glovebox atmosphere, the
material was stored in a glass ampoule clamped shut with a ground glass lid.
Rubidium T elluride, szT e — In an N; filled glovebox, 3 500g ampoule of
Rb metal was heated to 80°C in an oil bath. Once the Rb metal was molten,
13.08g (0.15 mol) was transferred to a 1000 mL three neck round bottom flask
The two side necks were closed off with ground glass stoppers, and the center
neck was connected to a glass adapter with a stopcock joint. The flask and adapter
was removed from the glovebox, connected to a condenser apparatus, and chilled
to -78°C using a dry ice/acetone bath. Approximately 400mL of NH3 were

condensed onto the Rb metal, under an N; flow, giving a dark blue solution. One

36

 

\"fiJ .

...A.

‘3‘ ‘1

‘1‘ 0')!
it hL
l

\-

‘.u

‘O

of the side arm stoppers was gently removed and a Teflon coated magnetic stir bar
was added to the solution, followed by 17.52g (0.21 mol) of Te. The glass stopper
was replaced and an additional 400mL of NH; was condensed into the flask From
this point on, the reaction proceeds as described above for KzTe. A bright yellow
powder resulted which was evacuated on a Schlenck line for 5 hours, taken into a
N2 filled glovebox, ground to a fine powder, and stored in the same manner as for
KzTe.

Cesium Telluride, CszT e -— The procedure was the same as desribed above
for szTe. Amounts of 20.07g (0.15 mmol) Cs and 9.63g (0.07 mmol) were used,
which represent a slight excess of Cs and a slight deficiency of Te. A bright
yellow powder resulted which was taken into the glovebox and stored in the same

manner as for KzTe.

2. Synthesis - All manipulations were carried out under a dry nitrogen
atmosphere in a Vacuum Atmospheres Dri-Lab glovebox.

CsCe3Te3 (I) —- Initial investigations into the CszTe/Cu/Ce/I‘e system
produced the ternary compound, CsCe3Te3, as a result of phase separation in
which the Cu did not incorporate into any of the Ce-containing products.
Amounts of 0.393g CszTe (1.0 mmol), 0.032g Cu (0.5 mmole), 0.070g Ce (0.5
mmol), 0.383g Te (3.0 mmol) were weighed into a vial in an N2 filled glovebox.
The reactants were thoroughly mixed and loaded into a 9 mm silica ampoule. The

atripoule was removed from the glovebox, evacuated on a Schlenck line to less

37

C-‘ I

*"'

I... vn

v 4"
in hit

Ls.

‘hdn...
: .

a,
uh ‘\
""

 

than 2.0 x 10“ mbar, and flame sealed. The reactants were heated to 550°C for 4

days, cooled to 100°C at 4°C/hr, and quenched to 50°C in 4 hours. The ampoule

was opened with a glass cutter and placed into a 100 mL tube containing a side

arm attachment to allow for the purging of N2 and filled with degassed DMF. As
the excess CszTex flux dissolved in the DMF, the solution turned a dark purple
color. Successive portions of degassed DMF were added until the solution
remained clear. The product was washed with ether and dried under a constant
flow of N2. The remaining material consisted of a red-brown powder as the major
phase and black hexagonal-shaped plate crystals as a minor phase. The identity of
the red-brown powder was confirmed by EDS to be CeTe3 while the hexagonal
crystals analyzed as ternary having an average composition of CSLoCCijng.

The reaction conditions were then optimized to produce CsCe3Te3 as a
major phase by removing the Cu from synthesis, raising the reaction temperature,
and using stoichiometric amounts of the reactants. The optimized reaction mixture
consisted of 0.079g CszTe (0.2 mmol), 0.168g Ce (1.2 mmol), and 0.383g Te (3.0
mmol) which was heated to 850° C for 5 days, cooled to 400°C at 4°C/hr and
100°C at 10°C/hr, and quenched to room temperature in 1 hour. Although these
conditions showed significant improvements in forming the desired product, they
still did not yield a pure product. However, the small amount of CeTe; powder
Which resulted could be removed from the crystals by simply sonicating the

mixture. The identity of the hexagonal crystals was confirmed by comparing the

38

r. 15‘
‘3‘».

i
V "‘
i.‘.'..

ii:' .‘Ii

u ..
T. -

w --u\\

,2

92¢
ail

‘i'

3“

» I~
‘V
.5“:

Lr’

 

, t
. l
.,‘L
, \
‘t
{-u
“H
n.
i‘ I

   

 

 

 

 

powder X-ray diffraction pattern of the product against one calculated using single
crystal X-ray data (see Table 2.1).

RbCe3Te3 (II) -— The synthetic route to the formation of RbCC3TCg was
very similar to that of CsCe3Te3. The compound was initially discovered from
reactions in the szTe/Cu/Ce/l" e system and fiirther Optimization was needed to
produce the pure compound. The optimized reaction consisted of amounts of
0.358g szTe (1.2 mmol), 0.126g Ce (0.9) mmol), and 0.612g Te (4.8 mmol) that
were weighed into vial in an N; filled glovebox, thoroughly mixed, and loaded
into a 9 mm silica ampoule. The ampoule was removed from the glovebox,
evacuated on a Schlenck line to less than 2.0 x 10‘1 mbar, and flame sealed. The
reactants were heated to 400°C in 12 hours, isothenned at this temperature for 12
hours, raised to 850°C in 22 hours, and isothermed at this temperature for 6 days.
The reaction was then cooled to 400°C at 4.5°C/hr followed by quenching to 50°C
in 4 hours. The product was isolated in the manner described above for CsCe3Te3.
The remaining material consisted of a small amount of red-brown powder while
the major phase was copper-colored hexagonal shaped plates. Typical yields were
26%, based on Ce. The identity of the red-brown powder was confirmed by EDS
to be CeTe3 while the hexagonal crystals analyzed as ternary having an average
composition of RbLoCeuTe. 1.2- The identity of the hexagonal plates was
confirmed by comparing the powder X—ray diffraction pattern of the product

against one calculated using single crystal X-ray data (see Table 2.2).

39

.0

..‘n'. - 'I

I“.

*e k.

”‘95

in

.u'
‘4. V‘l‘ "\

 

 

 

KC€3T88 (III) — A mixture of 0.309g KzTe (1.5 mmol), 0.126g Ce (0.9
mmol), and 0.612g Te (4.8 mmol) was thoroughly mixed in a scintillation vial in
an N; filled glovebox and loaded into a 9 mm silica ampoule. The ampoule was
removed from the glovebox and evacuated on a Schlenck line to < 2.0 x 10‘4 mbar
and flame sealed. The reactants were heated to 400°C in 12 hours, isothermed at
this temperature for 12 hours, raised to 850°C in 22 hours, and isothermed at this
temperature for 6 days. The reaction was then cooled to 400°C at 4.5°C/hr
followed by quenching to 50°C in 4 hours. The product was isolated in the
manner described above for CsCe3Te3. The remaining material consisted of a
small amount of red-brown powder while the major phase consisted of copper
colored hexagonal-shaped plate crystals. Typical yields were 41%, based on Ce.
The identity of the red-brown powder was confirmed by EDS to be CeTe3 while
the hexagonal crystals analyzed as ternary having an average composition of
KLOCelsTegg. The identity of the hexagonal crystals was confirmed by comparing
the powder X-ray diffraction pattern of the product against one calculated using
Single crystal X-ray data (see Table 2.3).

ICNd3Te8 (110 - The synthetic route to the formation of KNd3Te3 was very
similar to that of ACe3Te3 (A = Cs, Rb). The compound was initially discovered
from reactions in the K2Te/Cu/N dfT e system and fiirther optimization was needed
to produce the compound pure. The Optimized reaction consisted of amounts of
0.309g KzTe (1.5 mmol), 0.130g Nd (0.9 mmol), and 0.612g Te (4.8 mmol) which

were weighed into a vial in an N2 filled glovebox, thoroughly mixed, and loaded

40

" ’3

'Id

r

[I

I“; '31
9A .h

 

 

 

 

 

 

 

 

 

 

 

into a 9 mm silica ampoule. The ampoule was removed from the glovebox and
evacuated on a Schlenck line to < 2.0 x 104 mbar and flame sealed. The reactants
were heated to 400°C in 12 hours, isothermed at this temperature for 12 hours,
raised to 850°C in 22 hours, and isothermed at this temperature for 6 days. The
reaction was then cooled to 400°C at 4.5°C/hr followed by quenching to 50°C in 4
hours. The product was isolated in the manner described above for CSCC3TCg.
The remaining material consisted of a small amount of greenish-grey powder
while the major phase consisted of silvery, black hexagonal-shaped plate crystals.
The identity of the greenish— grey powder was confirmed by EDS to be NdTe3
while the hexagonal crystals analyzed as ternary having an average composition of
KLoNdleTeéh. The identity of the hexagonal crystals was confirmed by comparing
the powder X-ray diflraction pattern of the product against one calculated using

single crystal X-ray data (see Table 2.4).

41

Till:

Table 2.1 Calculated and Observed X-ray Powder Diffraction Pattern for
CsCe3Te8 (1)

 

 

hkl data) data) III...(obs)(%)
0 0 1 14.6676 14.9437 1.19
0 0 2 7.3338 7.4007 3.14
0 0 3 4.8892 4.9182 2.49
0 0 4 3.6669 3.6817 25.16
1 3 2 3.3522 3.3512 1.20
0 4 0 3.2490 3.2672 1.80
1 4 0 3.0541 3.0576 4.40
0 0 5 2.9335 2.9425 100.00
0 0 6 2.4446 2.4505 10.82
1 2 -6 2.2965 2.2922 0.55
0 0 7 2.0954 2.0989 1.95
0 0 8 1.8334 1.8361 5.85
l 1 -9 1.6350 1.6315 0.66
3 5 -6 1.6096 1.6130 0.45

 

 

 

42

Table 2.2 Calculated and Observed X-ray Powder Diffraction Pattern for

 

 

 

 

 

 

RbCe3Te8 (11)
h kl daifi) dflm) my (obs) (%)
001 14.1852 15.48415 7.6
1 10 7.3653 7.26646 33.8
0 o 3 4.7284 4.75831 13.1
2 0 -2 4.0800 4.07829 6.8
0 0 4 3.5463 3.54577 100.0
2 2 2 3.1086 3.10736 2.4
2 3 -2 2.9704 2.96543 56.3
o 0 5 2.8370 2.83265 94.1
2 3 2 2.7413 2.74177 35.2
2 0 -5 2.5852 2.58091 7.3
2 3 .4 2.4761 2.47396 23.4
o 0 6 2.3642 2.36725 68.8
0 6 0 2.1665 2.15922 6.3
2 3 -7 1.7894 1.78768 3.0
4 0 4 1.7698 1.76979 6.3
4 3 -5 1.7450 1.74353 13.5
0 6 5 1.7219 1.72274 4.3
5 21 1.6810 1.68111 2.6
2 6 -5 1.6605 1.65617 7.0
4 3 -6 1.6369 1.63680 9.1
5 3 1 1.6149 1.61386 13.9
5 3 2 1.5604 1.56038 12-5

 

\

43

 

 

Table 2.3 Calculated and Observed X-ray Powder Diffraction Pattern for
KCe3Te8 (111)

Md dentin and) III... (obs) ("4)
0 1 1 9.4636 8.3925 18.8
0 0 2 6.8921 6.9301 5.0
0 0 3 4.5947 4.5455 6.6
0 0 4 3.4461 3.3915 73.9
0 4 0 3.2540 3.2054 20.3
0 3 3 3.1545 3.1430 6.9
2 1 3 2.8987 2.8947 5.3

2 3 2 2.7249 2.7123 94.6
2 4 0 2.6311 2.6321 3.9
2 3 3 2.4527 2.4660 3.1
3 2 2 2.4048 2.3995 4.5
4 0 -1 2.2657 2.2650 12.6
4 0 0 2.2359 2.2360 3.6
061 2.1430 2.1420 11.1
0 1 7 1.9470 1.9449 100.0
3 5 3 1.8770 1.8760 42.4
423 1.8182 1.8168 3.2
2 3 6 1.7566 1.7555 1.9
1 2 8 1.5933 1.5898 9.3

 

   

Table 2.4

Calculated and Observed X—ray Powder Diffraction Pattern for

 

 

1(Nd3T'e8 (IV)

hkl data) and) III... (obs) We)
0 0 1 13.6692 12.95992 19.2
0 2 1 5.8095 5.80944 4.4
2 1 1 3.8325 3.82916 30.6
0 4 0 3.2090 3.21164 100.0
3 1 -1 2.9039 2.90268 77.7
1 3 3 2.8429 2.85832 29.3
1 4 -2 2.8182 2.82570 30.8
3 2 1 2.5561 2.55832 10.7
1 2 5 2.3285 2.33094 18.1
0 6 0 2.1665 2.20959 14.5
061 2.1136 2.11526 5.0
1 6 -1 2.0675 2.06435 11.0
4 1 -4 1.9869 1.99423 20.9
4 0 3 1.8710 1.86784 16.9
1 6 -3 1.8208 1.82325 59.0
2 7 0 1.6936 1.69531 23.5
1 4 -7 1.6826 1.67759 3.9
02 8 1.6511 1.64719 24.9
4 3 -6 1.6070 1.60734 7.7

 

45

a. "\r“
we, ,
t. .

‘1
,-
:3!
ED

>4 its

.' “Nu

3. Physical Measurements
Semiquantitative Energy Dispersive Spectroscopy (EDS) - The analyses
were performed using a JEOL JSM-6400V scanning electron microscope (SEM)
equipped with either a Noran TN—5500 or a Noran Vantage energy dispersive
spectroscopy (EDS) detector, depending on when the data were collected. Data
were acquired on several crystals using an accelerating voltage Of 25 kV and 40

sec accumulation time.

Powder X-ray Drfii'action - Analyses were performed using a calibrated
Rigaku Rotoflex rotating anode powder diffractometer controlled by an IBM
computer and Operating at 45 kV/ 100 mA with a 1°/min scan rate, employing Ni-
filtered Cu radiation. Samples were ground to a fine powder and mounted by
spreading the sample onto a piece of double sided scotch tape affixed to a glass
slide. Powder patterns were calculated using the Cerius2 software.18

X-ray Crystallography - CsCe3Te8 and RbCe3Te3; A single crystal of each
was mounted on the tip of a glass fiber. Intensity data were collected on a Rigaku
AF C6S four-circle automated diffractometer equipped with a graphite-crystal
monochromator. The unit cell parameters were determined from a least-squares
refinement using the setting angles of 20 carefully centered reflections in the 8° 3
20 S. 30° range. The data were collected with an 03-20 scan technique over one-
quarter of the sphere of reciprocal space, up to 60° in 20. Crystal stability was

monitored with three standard reflections whose intensities were checked every

46

“1 1
1.
“a vi
a. .

i
5 ; ...
1.1.x

W
..i ‘N
1 \
1..) _..
1
-. a.
1\ ‘
1. .....
r

 

 

150 reflections. No significant decay was detected during the data collection
period. An empirical absorption correction based on ‘I’-scans was applied to all
data during initial stages of refinement. A DIFABS19 correction was applied after
full isotropic refinement, after which fiill anisotropic refinement was performed.
The structures were solved by direct methods using the SHELXS—8620 package of
crystallographic programs and full matrix least squares refinement was performed
using the TEXSAN software package”. Crystallographic data for these
compounds are given in Table 2.5.

KCe3Te8: A single crystal was mounted on the tip of a glass fiber.
Intensity data were collected at 173.1K on a Siemens SMART Platform CCD
diffi'actometer using graphite monochromatized MO K01 radiation. The data were
collected over a full sphere of reciprocal space, up to 50° in 20. The individual
frames were measured with an 0) rotation of 03° and an acquisition time of 40 sec.
The SMART22 software was used for the data acquisition and SAINT 23 for the
data extraction and reduction. The absorption correction was performed using
SADABS.24 The structure was solved by direct methods using the SHELXTL25
package of crystallographic programs. The complete data collection parameters
and details Of the structure solution and refinement is given in Table 2.5.

KNd3Te3: A single crystal was mounted on the tip of a glass fiber,
Intensity data were collected on a Nicolet P3 four-circle automated diffractometer

equipped with a graphite-crystal monochromator. The unit cell was determined by

47

|
nun.“-

l
0‘. h'

.
..J'» 1...
\

'Q“.«_,
n UN ,1

taking a rotational photo Of the crystal and selecting 15-20 reflections from the
resulting fihn. These reflections were manually centered and indexed, a least
square refinement was performed, followed by a unit cell transformation to give
the highest symmetry cell. The data was collected over one-quarter of the sphere
Of reciprocal space, up to 60° in 20. The structure was solved in the same manner
as described for CsCe3Te3 and RbCC3TCg and the complete data collection
parameters and details Of the structure solution and refinement is given in Table
2.5.

Transmission Electron Illicroscopy - Electron difli'action studies were
carried out on a JEOL 100CX transmission electron microscope (TEM) using an
electron beam generated by a CeB6 filament and an acceleration voltage of 120
kV. After the samples were ground to a fine powder in acetone, the specimens
were prepared by dipping a carbon-coated grid in the suspension. The samples
showed no decomposition under the electron beam.

Magnetic Susceptibility Measurements - The magnetic response of the
compound was measured over the range of 2-300K using an MPMS Quantum
Design SQUID magnetometer. Samples were ground to a fine powder to
minimize anisotropic effects, and corrections for the diamagnetism of the

compound and PVC sample containers were applied. Magnetic susceptibility as a
function of field strength (at a constant temperature of 300K) was first investigated
to determine if the samples experienced saturation of their magnetic signal. For all

Samples, the magnetization increased linearly with increasing field over the range

48

met"!

1.. Mt.

.\l‘ "q

.{t t \Ill

H'ft
..h.‘ e

3

r ‘\
“bk .4,

an.

1
N“ M 1.

W”
..‘K

r2.

(1.
D

r

!._r .
('1‘.
{—‘r-

 

 

 

investigated (0-10,000G). Subsequent temperature-dependent studies were then

performed at a constant field. From the temperature dependent data, the molar
magnetic susceptibility, xM, was calculated and a plot of UM vs T was used to

derive the effective magnetic moment, peg, fiom the following formula:
__ 1/2
“eff— 2.818 x (x...)

where m is the inverse of the slope taken from the linear region Of the plot.
Charge Transport Measurements - DC electrical conductivity and
thermopower studies were performed at room temperature. Conductivity
measurements were performed in the usual four-probe geometry with 60- and 25-
mm diameter gold wires used for the current and voltage electrodes, respectively.
Measurements Of the sample cross-sectional area and voltage probe separation
were made with a calibrated binocular microscope. Conductivity data were
Obtained with the computer-automated system described elsewhere.26
Thermoelectric power measurements were made by using a slow AC technique27
which requires the production of a slowly varying periodic temperature gradient
across the samples and measuring the resulting sample voltage. Samples were
suspended between quartz block heaters by 60-mm gold wires thermally grounded
to the block with GE 7031 varnish. The gold wires were used to support and
conduct heat to the sample, as well as to measure the voltage across the sample
resulting from the applied temperature gradient. The magnitude of the applied

temperature gradient was generally 1.0K. Smaller temperature gradients gave

49

,0

r);
t.

90"-

‘u- 11

.,.
[a

 

essentially the same results but with lower sensitivity. In both measurements, the

gold electrodes were held in place on the sample with conductive gold paste.

Mounted samples were placed under vacuum (10.3 Torr) and heated to 320 K for
2-4 h to cure the gold contacts. For a variable—temperature run, data (conductivity
or thermopower) were acquired during sample warming. The average temperature
drift rate during an experiment was kept below 0.3 K/min. Multiple variable-
temperature runs were carried out for each sample to ensure reproducibility and

stability. At a given temperature, reproducibility was within i 5%.

50

11bit

 

 

Table 2.5 Crystallographic Data for ALn:,Te8 (A = Cs, Rb, K; Ln = Ce, Nd)

 

 

Formula CsCe;,Te8 RbCe3Te8
a, (A) 9.057(2) 9.051(2)
b, (A) 12.996(3) 12.996(3)
c, (A) 14.840(3) 14.376(3)
13, (deg) 9874(2) 9887(2)
v, (A ) 1726.4(7) 1670.8(7)
Space Group P21/a (#14) P21/a (#14)
Z value 4 4
F.W (g/mol) 1574.06 1526.63
d...“ (g/cm3) 6.056 6.069
u, (cm‘l) 232.41 246.98
crystal (mm3) 0.18x0.27x0.09 0.23x0.45x0.02
Radiation Mo K01 MO K0.
29...... (deg) 50.0 50.0
Temp., (°C) 293 293
NO. data collected 3412 3301
NO. unique data 3193 3096
NO. F02>3o (F02) 1591 1565
No. variables 110 l 10
R/Rw, % . 4.9/6.3 6.8/7.9
GOF 2.29 3.12

 

TR=Z(IF,|-IF,l)/ZIF,I Rw={2(W(lFol-chlf/ZWIFJZV’Z

51

ililit

Table 2.5 continued Crystallographic Data for ALn3Te8 (A =Cs, Rb, K; Ln = Ce,

 

 

Nd)
Formula KCe3Te8 KN d3Te8
a, (A) 9.0630(3) 8.956(1)
b, (A) 13.0164(4) 12.836(2)
c, (A) 13.9677(3) 13.856(3)
13, (deg) 99.305(1) 9942(1)
v, (A ) 1626.05(8) 1571.4(8)
Space Group P21/a (#14) P21/a (#14)
Z value 4 4
PW (g/mol) 1480.26 1492.62
d... (g/cm3) 6.047 6.308
.11, (cm") 225.40 246.46
crystal (mms) 0.18x0.18x0.02 0.18x0.21x0.03
Radiation MO KOt MO Ka
20m, (deg) 50.0 50.0
Temp., (K) 173 188
No. data collected 3769 2373
No. unique data 3768 2200
NO. F02>3o (F02) 3270 654
NO. variables 1 10 l 10
R/Rw, % “ 9.4/11~4 5.4/6.2
GOF 2.20 2.56

 

aR=2(|F.|-|F.|)/ZIF.| R.={lellF.1-IF.I)2/>3wIF.IZ}‘°

52

S. :11

1 .
”a ‘I.’
“‘5 a...

517
S.

I! r“,

“kl

3‘,”_
"\..i

C. Results and Discussion

Structure Description - The four isostructural compounds, ALn3Te8 (A =
Cs, Rb, K; Ln = Ce, Nd), resulted from initial investigations into the A/Cu/Ln/T e
(Ln = Ce, Nd) systems. Their two—dimensional structure is shown in Figure 2.1.
The Ln and Te atoms make up the anionic layers and the alkali cations reside in
the interlayer gallery. The Ln atoms possess three crystallographic positions with
two distinct coordination environments, shown in Figure 2.2. Two of the Ln
atoms are eight coordinate with a bicapped trigonal prismatic environment Of Te.
The third Ln atom is nine coordinate with a tricapped trigonal prismatic
environment of Te. The atomic coordinates and isotropic displacement
parameters are given in Table 2.6 and the anisotropic displacement parameters in

Table 2.7 for ALn3Te8 (A = Cs, Rb, K; Ln = Ce, Nd).
The anionic layer of these compounds is a derivative of the NdTe3 structure
type, differing only in the occupancy Of the square Te net. From a fully occupied

NdTe3 lattice, one tellurium atom is removed, causing the remaining Te atoms to

"condense" into Te32' oligomers and zig-zag (Te?)n polymers arranged in an
unusual pattern, shown in Figure 2.3. The bonding in the zig-zag chains consists

of almost equal Te-Te distances of 2.989(3)A and 3.010(3)A for CsCe3Te8. The

Te-Te distances in the trimerized ref unit are 2.836(3)A and 2.847(3)A, longer

than the normal Te-Te bond length of 276.4 found in (1>h,1>),re,.28 The formal

OXidation states are therefore A+(Ln3Te3)3+(Te32-)(T622h)“. Selected distances and

53

-u, e.

14.34;

“ear:
.l. H
‘

.

for. ‘
"I‘M.

 

bond angels for ALn3Te8 (A = Cs, Rb, K; Ln = Ce, Nd) are shown in Table 2.8.
The structure Observed for the defect square net in ALn3Te8 can be thought of as 3
2a x 3h superstructure Of the NdTe3 structure, which is thought to have an ideal

square sublattice.

The pattern of the Te net in ALn3Te8 was previously predicted on
theoretical grounds by Lee and F oran15 in reporting the structure of RbDy3Se8,16

This compound was solved in a disordered model in the orthorhombic space group
Cmcm with a=4.0579(6)A, b=26.47(1)A, and c=3.890(9)A, but Weissenberg and
precession photographs indicated a very weak superstructure with asuper = 4am,
bsm = 3bsub, and em = cm. This superstructure could not be resolved

crystallographically. HOMO-LUMO energy calculations were made using Hi‘ickel

theory to predict the superstructure pattern of RbDy3Se8. Although the 2a x 3b
superstructure of ALn3Te8 is different from the 4a x 3b superstructure found in
RbDy3Se8, one of the two lowest energy patterns predicted for the Sc net in its

superstructure is depicted in the Te net Of ALn3Te8.

54

 

i . J /

 

 

t.

.4 _ .

 

 

 

 

 

 

 

 

 

 

 

Figure 2.1 ORTEP representation of the structure Of ALH3TCg as seen parallel to
anionic layer. (circles with nonshaded octants: A = Cs, Rb, K; large open circles:
Te; circles with shaded octants: Ln = Ce, Nd).

55

° 9 <9

. Teld Te3C

.. A A
c .49

Te2C

C82 TelC

 

 

 

 
 
  

 

’ a
e6a Te7 Te8a .

Te7b

b
Te5b Te4

F' -
lgure 2-2 A fragment of CSCC3T63 showing the coordination envrronment of
the CC atoms.

56

 

 

 

 

Figure 2. -

. . 3 View of the Te “net” of CSCC3T68 showing the Te32’ units and the

Infinite zi 2- .

h gzag (T92 )n chains. The shaded area indicates the unit cell of the
3P01hefi

is ca] parent structure of NdTe3. The Te square net in the NdTe; structure
’Of “We fully oecupicd.

57

F
i.
‘1
fin
n.
'1':
‘11
.

ilblt

“ohm
'4 t

Ain‘t.

Table 2.6 Fractional Atomic Coordinates and Equivalent Isotropic Displacement
Parameters (Beq) for ALn3Te8 (A = Cs, Rb, K; Ln = Ce, Nd) with Estimated
Standard Deviations in Parentheses.

 

 

 

 

 

 

CSCC3T83

_ atom x y z Beqa,A2
Ce(l) 0.5958(2) 0.4160(2) 0.8597(1) 068(6)
Ce(2) 0.9158(2) 0.5850(2) 0.1413(1) O.70(6)
Ce(3) 0.91 13(2) 0.2494(1) 0.1458(1) O.67(7)
Te(l) 0.8562(2) 0.5856(2) 0.9167(1) 071(7)
Te(2) 0.6478(2) 0.4199(2) 0.0844(1) 067(7)
Te(3) 0.8555(2) 0.2459(2) 0.9209(1) 071(8)
Te(4) 0.9434(2) 0.4195(2) 0.3148(1) 1.06(8)
Te(5) 0.7052(3) 0.2795(2) 0.3222(2) 122(8)
Te(6) 0.4631(2) 0.4206(2) 0.3144(1) l.05(8)
Te(7) 0.2033(2) 0.5987(2) 0.3121(1) 1.12(8)
Te(8) 0.0482(3) 0.2496(2) 0.6901(1) 1.0(1)
Cs(l) 0.2573(2) 0.4065(2) 0.5301(1) 216(9)

RbCe3Teg
atom x y z BeqasAz
Ce(l) 0.5942(4) 0.4159(2) 0.8551(2) 03(2)
Ce(2) 0.9167(4) 0.5849(2) 0.1466(2) 0.3(2)
Ce(3) 0.9124(4) 0.2496(2) 0.1514(2) 0.3(2)
Te(l) 0.8553(4) 0.5856(2) 0.9139(3) 040)
Te(2) 0.6486(4) 0.4200(2) 0.0871(3) 0.3(2)
Te(3) 0.8548(5) 0.2460(2) 0.9184(3) 0.4(2)
Te(4) 0.9466(5) 0.4195(3) 0.3258(3) 0.7(2)
Te(5) 0.7078(5) 0.2795(3) 0.3332(3) 1.1(2)
Te(6) 0.4662(5) 0.4204(3) 0.3256(3) 070)
Te(7) 0.2058(5) 0.5991(2) 0.3225(3) 1.3(2)
Te(8) 0.0450(6) 0.2494(3) 0.6788(3) 3.7%

W 0.2581(8) 0.4066(4) 0.5320(4) -()

 

8B YalueS for anisotropically refined atoms are given in the 2form of the2 isotropic
°§luwalent displacement parameter defined as 13,, = (4/3)[a 13(1, 1) + b B(2,2) +
c 3(3’3) + ab(9037)}3(1,2) + ac(cosB)B(l,2) +ac(cosB)B(1,3) + bc(coscl)B(2,3)]

58

13bit
h5g1;

Table 2.6 continued Fractional Atomic Coordinates and Equivalent Isotropic

 

 

 

 

 

 

 

'3 values for anisotropi
equivalent displacemen

2

Displacement Parameters (mg) for ALn3Te8 (A = Cs, Rb, K; Ln = Ce, Nd) with
Estimated Standard Deviations in Parentheses.
KCC3T83
atom x y z Ueqas A2
Ce(l) 0.5931(2) 0.4153(2) 0.8507(2) 044(1)
Ce(2) 0.9180(2) 0.5840(2) 0.1510(2) 044(1)
Ce(3) 0.9140(2) 0.2489(2) 0.1558(2) 044(1)
Te(l ) 0.8546(2) 0.5871(2) 0.9115(2) 044(1)
Te(2) 0.6489(2) 0.4220(2) 0.0895(2) 044(1)
Te(3) 0.8538(3) 0.2482(2) 0.9155(2) 044(1)
Te(4) 0.9495(3) 0.4177(3) 0.3357(2) 049(1)
Te(5) 0.7104(4) 0.2761(3) 0.3428(3) 055(1)
Te(6) 0.4679(3) 0.4206(3) 0.3357(2) 049(1)
Te(7) 0.2079(3) 0.6015(2) 0.3322(3) 054(1)
Te(8) 0.0426(3) 0.2503(2) 0.6683(2) 047(1)
._ K(1) 0.2590(1) 0.4051(8) 0.5358(9) 063(1)
qu is defined as one-third of the trace of the orthogonalized Ujj tensor.
KNd3Teg
atom x y z Beqas A2
Nd( 1) 0.6042(3) 0.4106(3) 0.8492(2) 0.8(1)
Nd(2) 0.9299(3) 0.5795(3) 0.1476(2) 0.8(1)
Nd(3) 0.9141(2) 0.254(1) 0.1548(2) 1.5(1)
Te(l) 0.8607(4) 0.5804(3) 0.9176(3) 1.4(1)
Te(2) 0.6544(4) 0.4143(4) 0.0933(3) 1.2(1)
Te(3) 0.8533(3) 0.245(1) 0.9167(2) 1.5(1)
Te(4) 0.9473(4) 0.4179(3) 0.3268(3) 1.6(1)
Te(S) 0.7102(4) 0.2777(2) 0.3411(3) 1.4(1)
Te(6) 0.4690(4) 0.4227(3) 0.3409(3) 1.5(1)
Te(7) 0.2076(4) 0.5967(3) 0.3303(3) 1.7(1)
Te(8) 0.0423(3) 0247(1) 0-6704(2) 17(1)
K(1) 0.259(1) 0.409(1) 0.5357(9) 24(4)

 

c 30.3) + ab(cosy)B(l

59

cally refined atoms are given in the form of the isotropic
t Parameter defined as 13.,q = (4/3)[a2B(l,l) + 1523(22) +
,2) + ac(cosB)B(1,3) + bc(cosct)B(2,3)]

11bit

 

 

Table 2.7

Anisotropic Displacement Parameters (A) for ALn3Te8 (A = Cs, Rb,
K; Ln = Ce, Nd) with Standard Deviations in Parentheses

 

 

 

 

 

CsCe3Teg

atom U11 U22 U33 U12 U13 U23
Ce(l) 0.0079(9) 0.0073(7) 0.010(1) 0.0001(9) 0.0010(7) 0.0001(8)
Ce(2) 0.0075(9) 0.0067(7) 0.012(1) 0.0001(9) 0.0006(7) 0.0006(8)
Ce(3) 0.008(1) 0.007(1) 0.010(1) 0.0002(8) 0.0011(8) 0.0008(8)
Te(l) 0.008(1) 0.0063(8) 0.013(1) 0.000(1) 0.0017(8) 0.001(1)
Te(2) 0.015(1) 0.011(1) 0.014(1) 0.001(1) 0.0017(9) 0.001(1)
Te(3) 0.007(1) 0.009(1) 0.010(1) 0.0003(9) 0.0011(9) 0.0002(9)
Te(4) 0.008(1) 0.008(1) 0.010(1) 0.000(1) 0.012(8) 0.0006(9)
Te(S) 0.013(1) 0.012(1) 0.015(1) 0.000(1) 0.022(9) 0.000(1)
Te(6) 0.012(1) 0.013(1) 0.018(1) 0.001(1) 0.0043(9) 0.003(1)
Te(7) 0.012(1) 0.024(1) 0.010(1) 0.000(1) 0.029(9) 0.003(1)
Te(8) 0.013(1) 0.011(1) 0.013(1) 0.0007(9) 0.001(1) 0.001(1)
g1) 0.030(1) 0.035(1) 0.017(1) 0.001(1) 0.005(1) 0.000(1)
Metres

Em 011 022 033 012 013 023
Ce(l) 0.004(3) 0.001(2) 0.007(2) 0.002(2) 0.003(2) 0.001(2)
Ce(2) 0.004(3) 0.001(2) 0.005(2) 0.003(2) 0.002(3) 0.001(2)
Ce(3) 0.003(3) 0.002(2) 0.007(2) 0.001(2) 0.004(2) 0.001(2)
Te(l) 0.005(3) 0.004(2) 0.006(3) 0.000(2) 0.002(3) 0.001(2)
Te(2) 0.002(3) 0.004(2) 0.005(3) 0.000(2) 0.002(3) 0.002(2)
T9(3) 0.004(3) 0.006(2) 0.005(3) 0.002(2) 0.001(3) 0.001(2)
T9(4) 0.009(3) 0.010(2) 0.008(3) 0.001(2) 0.002(3) 0.004(2)
Te(5) 0.004(3) 0.032(2) 0.007(3) 0002(2) 0002(3) 0.005(2)
T6(6) 0.010(3) 0.010(2) 0.009(3) 0.001(2) 0.004(3) 0.000(2)
Te(7) 0.017(3) 0.010(2) 0.021(3) 0.000(3) 0000(3) 0.002(2)
T48) 0.003(3) 0.006(2) 0.011(3) 0.001(2) 0.003(3) 0901(2)
Rb(1) 0.033(5) 0.035(3) 0.008(4) -0.002(4) -0.005(4) -0-002(3)

 

6O

 

Table 2.7 continued

Anisotropic Displacement Parameters (A) for ALn3Te8

(A = C8, Rb, K; Ln = Ce, Nd) with Standard Deviations in Parentheses

 

 

 

 

 

KCC3T68

atom U11 U22 U33 U12 U13 U23
Ce(l) 0.028(1) 0.045(1) 0.056(1) 0.006(1) 0.004(1) 0.003(1)
Ce(2) 0.030(1) 0.047(1) 0.053(1) -0.008(1) 0.002(1) 0.007(1)
Ce(3) 0.027(1) 0.051(1) 0.054(1) -0.009(1) 0.000(1) 0.005(1)
Te(l) 0.030(1) 0.044(1) 0.054(1) 0.004(1) 0.003(1) -0.008(2)
Te(2) 0.027(1) 0.044(1) 0.058(1) -0.003(1) 0.002(1) 0.004(1)
Te(3) 0.028(1) 0.056(1) 0.050(1) 0.004(1) 0.001(1) -0.006(2)
Te(4) 0.034(1) 0.058(1) 0.054(1) 0.007(1) 0.000(1) 0.001(2)
Te(S) 0.032(1) 0.101(2) 0.052(1) 0.002(1) 0.001(1) 0.010(2)
Te(6) 0.036(1) 0.055(1) 0.056(1) -0.007(1) 0.003(1) 0.001(2)
Te(7) 0.038(1) 0.034(1) 0.072(1) 0.000(1) 0.002(1) 0.003(1)
Te(8) 0.036(1) 0.044(1) 0.054(1) 0.006(1) 0.002(1) -0.001(2)
fl) 0.045(5) 0.062(5) 0.070(4) 0004(5) 0.003(5) 0.001(6)
Qd3Teg

film 011 022 033 012 013 023
MO) 0.005(1) 0.017(1) 0.010(2) 0.011(1) 0.008(1) 0.001(2)
Nd(2) 0.004(1) 0.019(1) 0.009(2) 0.009(1) 0.003(1) 0.003(1)
Nd(3) 0.027(1) 0.019(1) 0.010(1) 0.000(4) 0.003(1) 0.005(5)
Te(1) 0.024(2) 0.013(2) 0.013(3) 0001(2) 0.002(2) 0.006(2)
Te(2) 0.020(2) 0.034(2) -0.011(2) 0.004(2) 0.004(1) 0.001(2)
1‘96) 0.028(1) 0.024(2) 0.007(2) 0.004(5) 0.004(1) 0.010(5)
T9(4) 0.031(2) 0.032(2) 0004(2) 0003(3) 0.005(1) 0.002(2)
Te(5) 0.028(1) 0.015(2) 0.009(2) 0.000(1) 0.002(1) 0.003(1)
Te(6) 0.026(2) 0.016(2) 0010(2) -0.001(2) -0.008(1) 0.005(2)
Te(7) 0.028(1) 0.027(1) 0.009(2) 0.001(2) 0.005(1) 0.006(2)
Te(8) 0.029(1) 0.025(1) 0.011(2) -0.002(6) 0.002(1) 0.017(5)
1(0) 0.038(5) 0.035(5) 0.015(7) 0000(7) 0006(5) 0.016(7)

 

61

Table .

A I
Y), "1 '.
cut—A».

Table 2.8 Selected Distances (A) and Bond Angles (deg) for CsCe3Te8 with

 

 

Standard Deviations in Parentheses.
Cel — 'l‘el‘I 3.244(3) Cel — Te2' 3.296(3) Cel - T07a 3.356(3)
Ce2 - Te3" 3.249(3) Ce2 — Te3c 3.251(3) Ce2 — Te4b 3.336(3)
Ce2 - Te8‘ 3.278(3) Ce3 — Tc3' 3.300(3) Ce3 — Te4a 3.322(3)
Ce3 - Te6' 3.316(3) Te4 — Te5 2.836(3) Te4 — Te7 3.317(3)
Te4-Te8 4.277(2) Te5 -Te6 3.510(3) Te6—Te5 2.847(3)
Te6—Te7 3.297(3) Te6—Te8 4.284(2) Te7—Te8 2.988(3)
Csl — Te4 3.944(3) Csl - Te7 4.057(3) Csl — Te8 3.841(3)
Teld-Ce3-Te2“ 8835(7) Te1"-Ce2-Te2c 74.8 1 (6)
Telb-Ce3-Te3' 7408(6) Te1'-Ce2-Te4b 139.82(9)
Te3‘-Ce3-Te4‘ 139.05(8) Te4b-Ce2-Te7b 5946(6)
Te4"-Ce3-Te5a 4939(6) Te4b-Ce2-Te8’ 81.17(6)
Teld-Cel-Tez“ 7490(6) Te7’-Te8‘-Te7b 178.8(1)
Tel“-Ce1.re2c 8903(7) Ce3-Te2°-Cel 8270(6)
Te2‘-Ce1 -Te6‘ 139.27(9) Cel-Te3b-Ce2 8280(7)
Te6‘-Ce1-Te7' 5916(6)

 

62

Table I

gr I
\:';r a j!
.1“.

(ti

3‘
\o

’5
b.

Table 2.9
Standard Deviations in Parentheses.

Selected Distances (A) and Bond Angles (deg) for RbCe3Te8 with

 

 

63

Ce] —Te1‘I 3.247(6) Cel —Te2' 3.291(5) Cel —Te7° 3.357(7)
Ce2 — Te3" 3.253(6) Ce2 — Te3° 3.252(6) Ce2 — Te4b 3.332(5)
Ce2 — Te8‘ 3.282(5) Ce3 - Te3' 3.317(5) Ce3 - Te4' 3.317(5)
Ce3 - Te6' 3.316(5) Te4 — Te5 2.840(6) Te5 — Te6 2.842(6)
Te4—Te8 4.305(5) Te5 —Te6 3.509(5) Te6—Te5 2.841(4)
Te6—Te7 3.304(4) Te6—Te8 4.278(5) Te7—Te8 2.986(4)
Rbl — Te4 3.760(9) Rbl — Te7 3.886(7) Rb] — Te8 3.686(8)
Teld-Ce3-Te2° 88.1(2) Tel‘-Ce2-Te2° 74.7(1)
Telb-Ce3-Te3' 73.9(1) re1'-Ce2-Te4" 139.8(1)
Te3'-Ce3-Te4' 139.0(1) Te4b-Ce2-Te7" 595(1)
Te4'-Ce3-Te5' 496(1) Te4b-Ce2-Te8' 81.2(1)
Teld-Cel-TeZ" 74.8(1) Te7'-Te8'—Te7b 179.3(2)
Teld-Cel-Te2° 88.9(2) Ce3-Te2°-Ce1 82.8(1)
Te2'-Ce1-Te6‘ 139.3(1) Cel-Te3b-Ce2 82.9(2)
Te6'-Cel-Te7' 592(1)

Table .

 

Table 2.10 Selected Distances (A) and Bond Angles (deg) for KCe3Te8 with

Standard Deviations in Parentheses.

 

 

Cel — Teld 3.248(2) Cel — Te2' 3.294(2) Cel — Ter‘ 3.359(2)
Ce2 — Tc3b 3.256(2) Ce2 - Te3c 3.256(2) Ce2 — re4b 3.334(2)
Ce2 — Te8' 3.298(2) Ce3 - Te3‘ 3.307(2) Ce3 — Te4‘ 3.328(2)
Ce3 — re6' 3.322(2) Te4 — Te5 2.848(2) Te5 — Te6 2.850(2)
Te4—Te8 4.322(2) TeS —Te6 3.476(2) Te6 -Te5 2.882(2)
Te6—Te7 3.325(2) Te6-Te8 4.293(2) Te7 —Te8 2.973(2)
K1-T‘e4 3.641(7) K1-Te7 3.755(7) K1-Te8 3.549(7)
Tel"-Ce3-Te2° 88.13(5) Tel'-Ce2-Te2° 7454(5)
Telb-Ce3-Te3' 7404(5) Tel'-Ce2-Te4b 139.82(6)
Te3'-Ce3-Te4' 138.98(6) Te4b-Ce2-Te7b 5938(5)
Te4'-Ce3-Te5' 4958(5) Te4b-Ce2-Te8‘ 8098(5)
Teld-Cel-Te2' 74.80(5) _ Te7'-Te8‘-Te7" 179.7(1)
Teld-Cel-Te2° 8887(5) Ce3-Te2°-Cel 8278(4)
Te2'-Cel-Te6' l39.25(6) Cel-Te3b-Ce2 8284(5)
Te6'-Cel-Te7‘ 5906(5)

Table 2.11 Selected Distances (A) and Bond Angles (deg) for KNd3Te8 with

Standard Deviations in Parentheses.

 

 

Ndl — Tei‘I 3.196(5) Ndl — Te2“ 3.338(5) Ndl — Te7‘ 3.227(5)
Nd2 - Te3" 3.191(9) Nd2 - Te3c 3.321(8) Nd2 — Te4b 3.220(5)
Nd2 — Te8‘ 334(1) Nd3 — Te3' 3.257(4) Nd3 -— Te4a 3.159(9)
Nd3 — Te6' 341(1) Te4 — Tc5 2.815(5) TeS — Te6 2.851(5)
Te4 — Te8 4.302(4) Te5 — Te6 3.462(5) Te6 - Te5 2.851(4)
Te6—Te7 3.221(5) Te6—Te8 4.166(3) Te7—Te8 2.957(4)
K1 — Te4 3.68(1) K1 — Te7 370(1) Kl — Te8 357(2)
Teld-Nd3-Te2° 88. 1 (2) Te1'-Nd2-Te2° 75.6(1)
Telb-Nd3-Te3“ 73.4(2) Te1’-Nd2-Te4b 139.5(2)
Te3'-Nd3-Te4' 140.0(4) Te4"-Nd2-Te7b 606(1)
Te4'-Nd3-Te5' 506(1) Te4b-Nd2-Te8' 81.8(2)
Teld-Ndl-TeZ' 73.7(1) Te7'-Te8“-Te7b 178.9(4)
Teld-Ndl-TeZ" 87.9(2) Nd3-Te2°-Nd1 82.7(2)
Te2‘-Ndl-Te6‘ 139.3(2) Ndl-Te3b-Nd2 8296(8)
Te6'-Ndl-Te7’ 58.4(1)

 

 

65

1' .
'4 .,
Q‘gn‘l‘

Transmission Electron Microscopy - Electron diffraction studies on

KNd3Te8 revealed an additional, possibly incommensurate, superstructure along
the a-axis. The reflections associated with this new superstructure are very weak
and occur along the a' direction with aisuw = 0.429a*sub, where asub is the length
of the the KNdsTe8 cell (i.e., 8.956 A). An optical densitometric scan Obtained
from electron diffraction photographs Of the (hkO) reciprocal plane along the (h60)
row of reflections is shown in Figure 2.4. The weak reflections between the (160),
(260), and (360) reflections are due to the additional 0.429a"'sub superlattice, which
corresponds to a 2.33 asub (i.e.~ 21A) lattice dimension. These results suggest an

additional oligomerization and/or fiagmentation along the chains of the Te trimers

and/or the infinite zig-zag chains.

66

F0
:1: 2'4 (A) Selected are electron diffraction pattern of KNd3T68 With the
Perpendicular to the layers ([001] direction) (B) Densitometric intensity

scan ' -
along the b -aX1s of the electron diffraction pattern

67

Electron Diffraction of KNd 3Te 8

   

 

Intensity/Arbitrary Units

 

 

 

Reciprocal Angstroms

68

 

 

Magnetic Susceptibility Measurements — The magnetic susceptibilities Of
RbCe3Te8, KCe3Te8, and KNd3Te8 were measured over the range 5—300K at
50006, 60006, and 60006, respectively. A plot Of l/ltM vs T for each shows that

the materials exhibit nearly Curie-Weiss behavior with only slight deviation from

linearity beginning below 50K, see Figure 2.5. Such deviation has been reported
for several Ln3+ compounds and has been attributed to crystal field splitting of the
cation's 2F5,2 (Ce3+) and 419,2 (Nd3+) ground state.29 At temperatures above 150K, a
ueff of 2.7611B for RbCe3Te8, 3.0711B for KCe3Te8 and 3.3311B for KNdJTe8 has
been calculated. These values are in accordance with the usual range for Ce3+

(2.3-2.5%) and Nd3+ (3.5-3.62 ’03) compounds.

69

 

 

 

500 mmfim-wemnenqnn.
HA) 5

400? 0'1

. o .

i o . i

p; 300- o ..
I Q 1

\ l o . T
v-d zcx) L . l
E 0”. T

I . q

100 r/ 1

I l
0L4L1.JALIA--41--J-ILL-41...o-

 

0 50 100 150 200 250 300
Temperature (K)

350
300
250

E 200
150
100
50

O lLlJllJUlULllll+llJll

0 50 100 150 200 250 300

l/x

lllllllljllllljllljj A

 

mnluulnnn

 

 

 

Temperature(K)
250* -r.1-firlr...,....,...... ..
ol
2“) . . 1
. J
o 1
£150 . 0 '1
\ o i
100 , i
50 .0... i
i

 

() _L_non_mil - 1 141,1 1,. n n I n n n n I l n no; I n n n n
0 50 l 00 l 50 200 250 300
Temperature (K)

 

Figure 2.5 Inverse molar magnetic susceptibility (l/xM) plotted against
temperature (2-3OOK) for (A) RbCe3Teg, (B) KCesTes, and (C) KNdaTes.

70

Charge Transport Properties - Electrical conductivity data as a function Of
temperature for a single crystal Of KN dJTe8 and room temperature pressed pellets
Of CsCe3Teg and RbCC3TCg show that these materials are semiconductors with
room temperature values ranging from 0.01 to 0.1 S/cm, see Figure 2.6A. The
conductivities for CsCe3Teg and RbCe3Te3 decrease with decreasing temperature.
For KNd3TCg, however, the data do not follow the typical thermally activated
behavior of semiconductors, suggesting a complicated electronic band structure at
the Fermi level. Considering that grain boundaries in the pressed pellets of
CSCC3T83 and RbCe3Teg can inhibit their conductivities, it is difficult to deduce
which material is the most conductive. It can be said, however, that the pressed
pellets Of CsCe3Teg and RbCe3Teg are only slightly less conductive than the single
crystal of KNd3TCg. Also, the faster decline of the conductivity in the pellets at
lower temperatures can be attributed to thermal deactivation at the grain
boundaries.

Thermoelectric power data obtained on KngTeg and CsCe3Teg show a
very large Seebeck coefficient at room temperature of 500 and 400 uV/K,
respectively. The thermopower of RbCC3TCg has a much lower value of 200 tiV/K
at room temperature. The decreasing Seebeck coefficients with decreasing
temperatures, their positive signs, and their large magnitude (see Figure 2.6B)

confirm that these compounds are p-type, narrow gap semiconductors.

71

 

it 16
033511
‘ 7'

lit:
in): j

'Qt‘J'.

' 41“».

“IS. .

 

Considering that the structures of ALn3Te3 (A = Cs, Rb, K; Ln = Ce, Nd)
are related to that of the NdTe3 structure type, similar charge transport
measurements were made on a room temperature pressed Of the binary phase
CeTe3, for comparison. As shown in Figure 2.7A, the conductivity has switched
from that Of a semiconductor for ALn3Teg (A = C5, Rb, K; Ln = Ce, Nd) to that of
a metal for CeTe3 and the room temperature conductivity has increased by several
orders Of magnitude from 0.01-0.1 S/cm to 700 S/cm. In addition, the
thermopower data has decreased from 200-500 uV/K for ALn3Teg (A = Cs, Rb, K;
Ln = Ce, Nd) to 6-9 pV/K for CeTe; at room temperature. The thermopower data
Of CeTe3 neither follows the normal behavior for that Of a semiconductor nor a
metal and appears to reach a local minimum around 125K. It can be concluded
from this data that the structural differences imposed upon CeTe3 to form ALn3Te3

(A = Cs, Rb, K; Ln = Ce, Nd) have dramatically affected the physical properties.

72

A p-..J\To. v C

A Z\>-‘V cf.

0 Electrical Conductivity

 

 

 

   

 

 

10
(A) RI)C¢3T¢a
10-1 1(l‘ld3Tea
CsCe Te
104 .t’ 3 a
A a. /
§ 10'3 /
v 104
c /
10-5 ’4"
10"
10'7 LLLLLLLLanJJ1.14L1_LJI_LnInnnJ
0 50 100 150 200 250 300
Temperature (K)
Therm0power
(B) KNd-"Te8
400 :- / CsCeSTea
A
55 300 -
> .
1 C
v I
to 2m 1.- RbCeJTcs
100 :-
D.JJLJILJ;JLJLLIllllml__llj_¥1_
50 100 150 200 250 300

Temperature (K)

Figure 2.6 (A) Four probe, electrical conductivity [logo (S/cm)] and (B)
Thermopower (S/cm) data plotted against temperature (K) for room temperature
pressed pellets Of CsCe3Teg and RbCC3TCg and a single crystal of KngTCg.

73

Electrical Conductivity

12m I'I'ltijrrI‘ITr"l‘rr'ITjI'T—rii

 

 

 

 

 

 

 

 

 

1100 (A)
1000 _-
g. :
900 :.
$3 :
{’3 800:
b :"
700 _'-
Sm: IILIJJLJLI4IJLIMIIIAII‘ll—Ll_
0 50 100 150 200 250 300
Temperature (K)
10 Thermopower
T
l-
8 r-
r.
2 6 r
> i: r
:1. - .
v 4 '0 o
m It . .. Q
n 0.... .."O’
2 ' . 0...... ‘
’ 1
.ijJanAIIIIAmInALIIAJILHUL
0 50 100 150 200 250 300

Temperature (K)

Figure 2.7 (A) Four probe, electrical conductivity (S/cm) and (B) Thermopower
(llV/K) data plotted against temperature (K) for a room temperature pressed
Pellets of CeTe;

74

In}; 1
“View
V

4+ "
r .1 _
"ui\_\
P

14':
~KI“

TI? 1 l

‘K

 

D. Conclusions

TO conclude, a series of compounds having the formula ALn3Teg (A = Cs,
Rb, K; Ln = Ce, Nd) has been discovered which possesses a two-dimensional
structure very similar to that of the binary phase, LnTe3. The major difference is
that now an alkali metal has been inserted between the layers. In addition, a site
occupancy wave type distortion exists in the Te net of this compound, giving rise
to Te32- trimers and infinite zig-zag 0,622.)“ chains. From the charge transport
measurements, it is evident that these structural modifications have drastically
affected the physical properties Of the material. A gap has opened up at the Fermi
Level and the properties have switched fi'om metallic for CeTe; to semiconducting

for ALn3Teg (A = Cs, Rb, K; Ln = Ce, Nd).

From a chemical point of view, it would be interesting to see if there
is any relationship between the type of metal inserted between the layers and
the distortion in the Te net. To try and answer this question, many attempts
were made to synthesize a ternary compound in the BaanyTez system with
hopes that a similar compound would form with Ba2+ ions inserted between
the layers Of LnTe3. Unfortunately, a compound of this type could not be

synthesized. In fact, this system proved to be especially challenging as the
barium (as Ba or BaTe) did not incorporate into the product at all. In all

cases, phase separation was observed to BaTe and CeTe3. It would be

75

tk-fidT‘

l
|bll

pal“!

Lilhl

worth pursuing this system further since, to date, no BaanyTeZ phases are

known.

76

lltltrt

 

 

References

 

Erlander, M.; Hagg, G; Westgren, A. Ark. Kemi. Mineral. Geo]. 1935,
128(1), No. 1, 1-6.

Haneveld, A.J.K.; Jellinek, F. Rec]. T rav. Chim. Pays-Bas. 1964, 83, 776.

Boher, P.; Gamier, P.; Gavarri, J.R.; Hewat, A.W. J. Solid State Chem.
1985, 57, 343.

(a) Andress, K.R.; Alberti, E. Z Metallkd 1935, 27(6), 126. (b) Das, D.K.;
Pitrnan, D.T. Trans. Am. Inst. Min, Metall, Pet Eng. 1957, 209, 1175.

Ban, 2.; Sikirica, M. Acta. Crystallogr. 1965, 18, 594.
Zheng, C.; Hoffman, R. J. Am. Chem. Soc. 1986, 108, 3078.

(a) Brechtel, E.; Cordier, G.; Schafer, H.Z. Z Naturforsch., B: Anorg.
Chem, Org. Chem. 1979, 34, 251; Z Naturforsch., B: Anorg. Chem, Org.
Chem. 1980, 35, l; J. Less-Common Met. 1981, 79, 131. (b) Cordier, G.;
Eisenmann, B.; Schafer, H.Z. Z Anorg. Allg. Chem. 1976, 426, 205. (c)
Cordier, G.; Schafer, H.Z.‘ Z Natwforsch., B: Anorg. Chem, Org. Chem.
1976, 32, 383. (d) May, N.; Schiifer, H.Z. Z Naturforsch., B: Anorg.
Chem, Org. Chem. 1974, 29, 20. (e) Dorrscheidt, W.; Savelsberg, G.;
Sttihr, J.; Schiifer, HZ. .1 Less Common Met. 1982, 83, 269.

(a) LnNi2$i3 (Ln=Sc,U): Ya Kotur, B.; Bodak, O.I; Gladyshevski, E.I. Sov.
Phys. Crystallogr. 1978, 23, 101. (b) ScNiSi3: Ya Kotur, B.; Bodak, 0.1.;
Mys'kiv, M.G.; Gladyshevskii, E.I. Sov. Phys. Crystallogr. 1977, 22, 151.
(c) SmNiGe3: Bodak, O.I.; Pecharskii, V.K.; Ya Mruz, 0.; Yu Zarodnik,
V.; Vivits'ka, G.M.; Salamakha, P.S. Dapov. Akad. Nauk. Ukr. RSR , Ser.
B. 1985, 2, 36.

(a) Lin, W.; Steinfink, H.; Weiss, F. Inorg. Chem. 1965, 4, 877; Wang, R.;
Steinfink, H.; Bradley, W.F. Inorg. Chem. 1966, 5, 142. (b) Pardo, M.-P.;
Flahaut, J.; Domange, L.C.R. Bull. Soc. Chim. Fr. 1964, 3267. (c) Ramsey,
T.H.; Steinfmk, H.; Weiss, E. Inorg. Chem. 1965, 4, 1154. (d) Norling,
B.K.; Steinfink, H. Inorg. Chem. 1966, 5, 1488.

Cody, J.A; Ibers, J .A. Inorg. Chem. 1996, 35, 3836.

77

 

13

20

21

22

23

24

25

26

Zhang, X.; Li, J.; Foran, B.; Guo, H.-Y.; Hogan, T.; Kannewurf, C.R.;
Kanatzidis, M.G. J. Am Chem. Soc. 1995, 117, 10513.

Sheldrick, W.J.; Wachhold, M. Angew. Chem. 1995, 107, 490.; Angew.
Chem. Int. Ed Engl. 1995, 34, 40.

Kanatzidis, M.G. Angew. Chem. 1995, 107, 2281; Angew. Chem. Int. Ed
Engl. 1995, 34, 2109.

Savelsberg, G.; Schafer, H. Z Naturforsch, B. 1978, 33, 370.
Lee, S.; Foran, B. J. Am. Chem. Soc. 1994, 116, 154.
Foran, B.; Lee, S.; Aronson, M. Chem. Mot. 1993, 5, 974.

F eher, F. Handbuch der Praparativen Anorganischen Chemie: Brauer, G.,
Ed; Ferdinand Enke: Stuttgart, Germany, 1954, 280.

CERIUSz, Version 2.0, Molecular Simulations Inc., Cambridge, England,
1995.

Walker, N.; Stuart, D. Acta Cryst. 1983, 17, 42.

Sheldrick, G.M., in Crystallographic Computing 3; Sheldrick, G.M.;
Kruger, C.; Doddard, R., Eds.; Oxford University Press: Oxford, England,
1985, 175.

Gilmore, G.J., Appl. Cryst. 1984, 17, 42.

SMART: Siemens Analytical Xray Systems, Inc., Madison, WI, 1994.

SAINT: Version 4.0, Siemens Analytical Xray Systems, Inc., Madison WI,
1994-1996.

SADABS: Sheldrick, GM. University of GOttengen, Germany, to be
published.

Sheldrick, GM. SI-IELXTL, Version 5; Siemens Analytical Xray Systems,
Inc.; Madison, WI, 1994.

Lyding, J.W.; Marcy, H.O.; Marks, T.J.; Kannewurf, C.R. IEEE Trans,
Instrurn. Meas. 1988, 37, 76.

78

 

27

28

29

Marcy, H.O.; Marks, T.J.; Kannewurf, C.R. IEEE Trans. Instrum. Meas.
1990, 39, 756.

Huflinan, J.C.; Haushalter, R.C. Z. Anorg. Allg. Chem. 1984, 518, 203.

Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements; Pergamon
Press: New York, 1984; p 1443.

79

Chapter 3
Structure and Properties of ACuUTe, (A = Cs, Rb, K):

A Comparison with KCuUSe3

80

1. TM

TE

L“

.21-
K;
(J

Money.
{Ii '
nucl‘n

$663
.‘,
,1“
‘1‘.

We“.
. .

 

A. Introduction

Recently, there have been several reports in the literature marking the
discovery of new members of a large family Of compounds with the general
formula, AMLnQ3 (A = alkali metal; M = coinage metal; Ln = Group IV or rare
earth metal; Q = chalcogenide). Approximately 30 new compounds have been
synthesized with this stoichiometry. However, they do not all adopt the same
structure type. There are four known structure types that exist for this
stoichiometry, see Figure 3.1. Nevertheless, the majority of these compounds do
crystallize in one structure type [type a] and therefore this will be the focus of this
chapter. The details of the remaining three structures (types b—d) can be found

elsewhere"8 (also described in Chapter 1).
Members of the AMLnQ; family that crystallize in this [type a] structure
type include NaCquS3,l’2 KCuUSe3,2’3 KCquQ3 (Q = 8, Se, Te),2’4 KCquS3,4
CsCuCe83,2’3 CsCuUTe3,2’5 BaCuLnS3 (Ln = Sc, Y, Gd, Er),6 BaCuLnSe3 (Ln = Y,
En,“ BaCuDyTe3,7 BaCuLnTe3 (Ln = Y, La, Pr, Nd, Yb),8 BaAgNdS36 and
BaAgLnSe; (Ln = Y, La, Er)°, BaAgLnTe3 (Ln = Y, La, 0d),8 and BaAquSe38.
While several of these compounds were solved from single crystal x-ray data,
many others were simply confirmed as isomorphous from their powder x-ray
diffraction patterns. In these cases, unit cells were determined either from a least
Square refinement on reflections indexed from these powder diffraction patterns or
from Weissenberg or precession photographs taken on single crystals. A

remarkable aspect of this structure is that it can be stablized for such a wide

81

'fl ‘Jm
Au .5.”

“A"
List. p

53“

mm.

 

—--- —A A —Lm._-_-___- '_,._ _, ,_, .._ .-.» . Lac-ao— I’m

variety of elements. It can form sulfide, selenide, and telluride analogs. This is
rare since most multinary tellurides tend to adopt completely different structure
types than the sulfides. The fact that there are no Q-Q bonds in this structure
makes this even more interesting.

Although most of these compounds have not been physically characterized
in terms of their charge transport properties, the data that is available suggest that
their properties are as varied as their elemental compositions. For example, in the
KCanQ3 (Q = Se, Se, Te) series,“ KCans3 is an insulator, KCquSe3 is a
semiconductor and KCquTe, is a metal. This suggests that the properties can be
tuned simply by varying the chalcogenide. The compound KCuUSe3 was
synthesized in our lab and was determined to be a semiconductor.2‘3 While the
thermopower at room temperature was very high, the electrical conductivity
seemed quite low. In order to optimize these properties for themoelectric
application, the electrical conductivity must be enhanced. We therefore decided to
make some isostructural tellurides and study their properties. The larger size of
the tellurium atoms are expected to narrow the bandgap and as a result, the
materials should possess a higher electrical conductivity. Here, we report the

synthesis, structure, and physicochemical properties Of ACuUTe; (A = Cs, Rb, K)

82

..——. ~7-

Ail“ E A V rd
. I a. ‘ \

 

 

Figure 3.1 The four structure types of AMLnQ3 (A = alkali 01' alkaline earth
metal, M = coinage metal, M’ =Group IV or rare earth metal, Q = chalcogenide).
(A) Cmcm structure (e.g.; KCquS3), (B) ana (1) structure (e.g.; NaCuTiS3), (C)
C2/m structure (e.g.; BaAgErsg), and (D) ana (11) structure (98-; BaCuLaSs).

The small open circles represent M, the black circles represent Ln, and the large
Open circles represent A and Q.

83

 

.. .
it" 1‘
AI“ 50
l,‘.

(‘7'! 1
\lt'V

 

 

B. Experimental Section

1. Reagents - The following reagents were used as obtained: Potassium
metal, analytical reagent, Spectrum Chemical Mfg. Corp., Gardena, CA;
Rubidium metal, 99.5%, Alfa Aesar, Ward Hill, MA.; Cesium metal, 99.98%,
Alfa Aesar, Ward Hill, MA; Copper metal, electrolytic dust, Fisher Scientific,
Fairlawn, NJ; uranium metal, 60 mesh, Cerac, Milwaukee, WI; Tellurium powder,
100 mesh, 99.95% purity, Aldrich Chemical Co., Milwaikee, WI.

Cesium T elluride, Cs zTe — Syntheses of these materials was performed as
described in Chapter 2, Section 8.1
Rubidium Telluride, szTe — Syntheses of these materials was performed
as described in Chapter 2, Section B.1.
Potassium Telluride, K zTe — Syntheses of these materials was performed

as described in Chapter 2, Section B.1.

2. Synthesis —All manipulations were carried out under a dry nitrogen
atmosphere in a Vacuum Atmospheres Dri-Lab glovebox.

ACuUTe; (A = Cs, Rb, K) (I-III) - Amounts of AzTe (0.5 mmol), Cu (1.0

0111101), U (1.0 rnmol), and Te (2.5 mmol) were weighed into a vial in an N; filled

glovebox. The reagents were thoroughly mixed and loaded into a 9mm carbon

coated silica ampoule. The ampoule was removed from the glovebox, evacuated

on a Schlenck line to less than 2.0 x 10" mbar, and flame-sealed. The reactants

were heated to 850°C in 12 hours, isothermed at that temperature for 6 days, and

84

cooled to room temperature at a rate of -4°C/hr. NO isolation was needed since the
compounds were prepared fi'om direct combination of the elements. The products
Obtained consisted of black microcrystalline powders in leO%. The identity of
CsCuUTe3 was confirmed by comparing the powder X-ray diffraction pattern of
the product against one calculated using single crystal X-ray data (see Table 3.1).
The identities of RbCuUTe3 and KCuUTe3 were confirmed by comparing the
powder X-ray diffiaction patterns of the products against ones calculated using the

atomic coordinates from the single crystal X-ray data for CsCuUTe3 with the new

unit cell parameters (see Tables 3.2-3.3).

85

Table L

[st

 

Table 3.1 Calculated and Observed X-ray Powder Diffraction Pattern for

 

 

CsCuUTe; (I)
hid it... (70 d...(A) III... (obs) 04)
0 2 0 8.3305 8.3814 100.0
0 2 2 4.6864 4.7054 14.96
1 10 4.1881 4.1921 17.55
0 2 3 3.4415 3.4331 24.34
1 1 2 3.3684 3.3778 9.82
l 3 1 3.2684 3.2870 28.62
1 3 2 2.9241 2.9393 16.33
0 0 4 2.8342 2.8451 15.63
1 1 3 2.8057 2.8184 8.44
0 6 1 2.6971 2.6946 7.93
1 5 0 2.6401 2.6567 12.13
0 6 2 2.4937 2.5096 18.10
1 5 2 2.3932 2.4070 21.48
1 1 4 2.3473 2.3553 12.49
0 6 3 2.2377 2.2507 3.89
1 5 3 2.1642 2.1759 10.25
2 2 0 2.0940 2.0981 19.47
2 2 1 2.0592 2.0640 5.77
2 2 2 1.9643 1.9693 16.96
2 4 1 1.8930 1.8978 9.54
2 2 3 1.8316 1.8375 11.94

 

86

Table

Tris.

 

Table 3.1 continued Calculated and Observed X-ray Powder Diffraction Pattern
for CsCuUTe; (I)

 

 

hkl daulA) dads) III... (obs) 0%)
0 6 5 1.7563 1.7661 4.16
0 4 6 1.7223 1.7291 7.26
2 6 1 1.6876 1.6900 4.44
262 1.6342 1.6441 6.12
0 6 6 1.5622 1.5647 3.81

 

87

Table
Tibial

 

 

 

-‘ Ahm“ '—.o

Table 3.2 Calculated and Observed X-ray Powder Diffraction Pattern for
RbCuUTe3 (II)

 

 

hkl data) and) III...(obs)(%)
0 2 0 8.3035 8.1762 100.0
0 O 2 5.6685 5.7700 14.05
0 2 2 4.6929 4.6944 14.79
1 l 1 3.9347 4.0455 13.02
0 2 3 3.4496 3.4538 28.25
1 3 0 3.4125 3.3973 46.75
1 3 2 2.9263 2.9183 36.39
0 0 4 2.8442 2.8656 22.04
1 1 3 2.8127 2.8301 15.68
0 2 4 2.6908 2.6903 15.09
1 5 1 2.5682 2.5925 15.09
1 5 2 2.3919 2.4272 25.89
1 1 4 2.3539 2.3608 34.76
0 4 4 2.3464 2.3313 13.91
1 3 4 2.1848 2.1839 19.38
2 0 2 2.0250 2.0101 40.24
0 0 6 1.8962 1.8913 25.44
1 1 6 1.7277 1.7329 11.83
2 4 4 1.5919 1.5898 12.28
1 9 3 1.5496 1.5461 12.87

 

 

88

 

Table 3.3 Calculated and Observed X-ray Powder Diffraction Pattern for
KCuUTe3 (III)

 

 

hkl «1.170 «1.148) mm (obs) W»)
0 2 0 7.6880 7.7533 100.0
0 0 2 5.6785 5.6998 19.06
0 2 2 4.6815 4.5880 12.51

1 1 1 3.9203 3.9266 11.17

1 3 0 3.3922 3.4073 31.24
0 4 2 3.3431 3.3186 53.18

1 3 2 2.9166 2.8671 52.64
0 0 4 2.8393 2.8430 35.32

1 l 3 2.8050 2.8054 12.24
0 6 0 2.7572 2.7421 11.57
0 2 4 2.6854 2.6703 16.72

1 5 1 2.5585 2.5759 11.17
1 3 3 2.5292 2.5173 16.05

1 1 4 2.3481 2.3473 40.13
0 0 5 2.3407 2.3021 37.79
1 5 3 2.1577 2.1680 18.19
1 7 0 2.0885 2.0908 16.52
1 5 4 1.9279 1.9324 18.26
2 4 1 1.8869 1.8767 13.04
1 7 3 1.8182 1.8283 29.10
1 1 6 1.7241 1.7240 11.24
0 2 7 1.5921 1.5918 10.57

 

89

for the

“ ‘

“8'3;
‘ n

uh. .\
v

3. Physical Measurements — The instrumentation and experimental setup
for the following measurements are the same as described in Chapter 2, Section
3.3: Powder X-ray Diffraction, Magnetic Susceptibility Measurements, and
Charge Transport Measurements.

Unit Cell Determinations — X-ray powder diffraction data was
collected on a Rigaku Denki/RW400F2 (Rotaflex) rotating anode X-ray
diifiactometer controlled by an IBM computer and operating at 45 kW 100 mA.
The data was collected with a l°/min scan rate from 2 to 60° 29, employing Nil-
filtered Cu radiation. A U-fit program9 was used to calculate new cell parameters
for RbCuUTe3 (II) and KCuUTe3 (111) from the known cell parameters for
CsCuUTe3 (I). The unit cell parameters for all three compounds are given in
Table 3.4.

Infrared Spectroscopy - Optical diffuse reflectance measurements were
made on a finely ground sample at room temperature. The spectrum was recorded
in the Mid-IR region (6000 — 400 cm") with the use of a Nicolet MAGNA-IR 750
Spectrometer equipped with a collector diffuse reflectance of Spectra-Tech Inc.
The measurement of diffitse reflectivity can be used to obtain values for the
bandgap which agree rather well with the values obtained by absorption
measurements from single crystals of the same material. Absorption (or! S) data

were calculated from the reflectance data using the Kubelka-Munk function10

90

3141.11

 

 

-.‘n—w;.ab -_ —— \

a/s = (1 — R)2/2R

where R is the reflectance at a given wavenumber, or is the absorption coefficient,
and S is the scattering coefficient. The scattering coefficient has been shown to be
practically wavenumber independent for particles larger than Sum, which is
smaller than the particle size of the samples used here. The bandgap was
determined as the intersection point between the energy axis at the absorption
offset and the line extrapolated from the linear portion of the absorption edge in

the 01/8 vs E(eV) plot.

91

Table

 

Table 3.4 Unit Cell Parameters for ACuUTe3 (A = Cs, Rb, K)

 

 

Formula CsCuUTe3 RbCuUTe3 KCuUTe3
a, (A) 4.3387(2) 4.3339(4) 4.3168(9)
b, (A) l6.664(2) 16.607(3) 16.543(3)
c, (A) 11.388(2) 11.377(1) 11.357(1)
v, (A3) 823.3(4) 818.9(4) 811.0(4)

Space Group Cmcm (#63) Cmcm (#63) Cmcm (#63)

92

 

. 3,.
Keith

5111':

‘- .‘2
54f“

 

C. Results and Discussion

Structure Description — The three isostructural compounds, ACuUTe3 (A =
Cs, Rb, K), are built from two basic building blocks; the [CuTe4] tetrahedra and
the [UTe6] octahedra. One-dimensional chains are formed when the Med
octahedra share edges down the a-axis, see Figure 3.2. These chains then comer
share at axial positions to form two-dimensional corrugated layers, see Figure 3.3.
These layers alone possess the anti-Pd3Te2 structure type in which Te atoms
occupy both the octahedral sites and the interlayer gallery positions.ll Within the
folds of these layers, however, the Cu atoms reside in tetrahedral sites, see Figure
3.4. It is interesting to note that the insertion of Cu at these sites does not add or
subtract fiom the dimensionality of the framework and the structure remains intact
upon removal of these copper atoms. The overall effect is a simple repeat pattern
of the two basic building blocks alternating across the layer. A view perpendicular

to the layers is shown in Figure 3.5.

93

Figur

" rs;
bin-‘5
s

Tim

 

 

Figure 3.2 Polyhedral representation of the one-dimensional chains built fiom
edge sharing connections of [UT e6] octahedrons in ACuUTe3 (A = Cs, Rb, K)

 

lfigure 3.3 Polyhedral representation of the two-dimensional corrugated layers
built from corner sharing connections of the one-dimensional chains in ACuUTe3

(A = Cs, Rb, K)

94

 

 

the

K) highlighting how

Rb,

= Cs,

“‘6 cepper atoms sit in the folds of the layers. The black circles represent U,
Striped circles represent Cu, and the open circles represent A and Te.
95

“Sum 3.4 Extended structure of ACuUTe; (A

' g}: '
9 o
. .I'OU

O

. ...lflfil.
O

0

Figure 3.5 View perpendicular to a single anionic layer of ACUUT€3 (A = C3,
Rb, K). The black circles represent U, the striped circles represent Cu, and the

Open circles represent Te.

96

.u,
‘I
i,

.

Jfia‘h

Shut
v

Tilt

‘1
Z“"'\
l ‘ |
“Nut

 

A M ‘ ‘J-‘h—Z (~- ,4 -I'-. -"-“ g ww

Magnetic Susceptibility Measurements - The magnetic susceptibilities of

CsCuUTe3, RbCuUTe3, and KCuUTe3 were measured over the range 5-300K at

3000G. A plot of 1/7CM vs T for each shows that the materials exhibit nearly Curie-
Weiss behavior above 140K, see Figure 3.6. Below this temperature, there is a
negative deviation from the straight line extrapolated from higher temperatures.
This phenomenon has been reported for several Ln4+ compounds and can be
attributed to crystal field splitting of the U4+ cation's ground state. At temperatures
above 140K, 3 p.eff of 3.4911,; for CsCuUTe3, 3.6911B for RbCuUTe3 for and 4.14111B
for KCuUTe3 has been calculated. These values are in accordance with what has
been theoretically calculated for a 5f2 U4+ ion (3.58 113)12 and experimentally
observed for several solid state transition metal/U‘Vchalcogenides (3.0-3.6 1.13)”.

From the data, it is evident that the compounds are valence precise and the
formula can be written as (AT)(Cu+)(UM)(Te2')3. We should therefore expect the

materials to exhibit semiconducting behavior.

97

 

 

 

 

 

 

 

 

 

 

 

350 b fir I— r I I I j I I j j j I 'j jfi I l I I I I I rT 1 1 .
t A E
300 F( ) . o . 1
O I
250 E' o . 3
i . ° 1
:5 'L . O O . 1
o ..
2 200 I ..O” . 1
I Q I
o .
150 E 1
I O l
100 if '5
I 1
50 hI L L. I I44_LLL LI_I_LI LJJJJ I I IJI LIJ‘
O 50 100 150 200 250 300
Temperature (K)
400 7 I I r I l I r fi—l' r. j r ii r I'_I rI I' I—I rI r h I—I fl 1
(B) -
350 . o J:
o O 1
300 . ’ 1
2 . ° ' i
5 250 o . . o 1
o .1
200 o... 1
° 1
150 . -
1
100 Jj L1 I ILLliLLJ II Ll LJ I I I I I I I I I I I I I I II
0 50 100 150 200 250 300 350
Temperature (K)
3% j 1T r. j I’ Ij I I I I—I I r I I I I I I I I l I I I I I I I i1:
280 L(C) o .3
I o .
260 L o ."
E e 1
240 _- ° -
g: ’3 o O :
2 220 E O . . 1
.. O 9 o 1
200 E'. ..O” 9 O .1
180 '- .‘ .1
I o 1
160 ; a
140 v _I .1 I I I I I4 L1 I I I I I I I I I I I I I I 14 I I_I I L I 1
0 50 100 150 200 250 300 350
Temperature (K)

Figure 3.6 Inverse molar magnetic susceptibility (llxM) plotted against
temperature (2-300K) for (A) CsCuUTe3, (B) RbCuUTei and (C) KCuUTes.

98

the t1

Charge Tranport Praperties - Both the electrical conductivity and
the thermopower was measured for hot pressed pellets (270°C) of ACuUTe3 and
the composite plots are shown in Figure 3.7. The electrical conductivity decreases
with decreasing temperature for all three compounds, indicating semiconducting
behavior with room temperature values ranging from 0.01 - 0.1 S/cm. It is
difficult to determine which of the three compounds is the most conductive,
however, since the measurement is very sample dependent and grain boundary
effects in the pellet act to artificially inhibit the conductivity. This is evident in
CsCuUTe3, where two measurements were made on two different pressed pellets
and the results are quite different. In comparison to KCuUSe3, whose
measurements were made on single crystals, the conductivity does not seem to
differ significantly. The room temperature value for this compounds was reported
to be 0.02 S/cm. However, if we assume that the conductivity of a pressed pellet
is ~100 times less conductive than a single crystal, the true conductivity of
ACuUTe; will be much higher than that for KCuUSe3. The thermopower
measurements, which are not affected by grain boundaries in the sample pellet
because it is a zero current technique”, give positive Seebeck coefficients at room
temperature of 100 uV/K for KCuUTe3 and 200 uV/K for CsCuUTe3. These
values are 3-6 times less than KCuUSe3 (600 uV/K at room temperature). This is
also an indication that the electrical conductivities of ACuUTe; are significantly
higher than that of KCuUSe3 since a decrease in thermopower is usually

accompanied by an increase in conductivity.

99

0 Electrical Conductivity

 

 

 

 

 

   

 

 

a.
- : b
(A) .
10'2 i d
e ' '
—4 -°
0 10 . -
\
U) .
v .
b 1045
.’ a - KCuUTe3 :1
-8 b -CsCuUT‘e3 ‘
10 . c -CsCuUTe3 i
. d - RbCuUT‘e3 ii
10-10 1 1 1 ,_L J 1 1 1 _l _L J L 1_l J #4 I i 1 1 4 1 l 1_i 4*;
O 50 100 150 200 250 300
Temperature (K)
Thermopower
250 _
2 " ' ~ L
00 C— CsCuUTe 4
3 1
2‘ L ‘
150 L -l
> r 1
1 ' .J
v E -1
U) 100 f _
,
l
.1
l
l

 

Temperature (K)

Figure 3.7 (A) Variable temperature, four probe electrical conductivity data for
hot pressed pellets of ACuUTe3 (A = CS, Rb, K)- (B) Variable temperature
thermopower data for hot pressed pellets of ACuUTe; (A = Cs, K).

100

 

[163$

 
 

A A - _.__-. A “v.5. “f ‘" ,“W—-‘I w-A— ,

Infi'ared Spectroscopy - The diffuse reflectance optical spectra were
measured in the Mid-IR region for all three compounds and are shown in Figure
3.8. From the charge transport properties, it is expected that these materials will
show an optical bandgap in this region. However, any optical bandgap present in
these region is masked by the presence of four peaks at 3511cm'l (0.43 eV),
161361111" (0.20 eV), 1349em'l (0.17 eV), and 898cm" (0.11 eV). These peaks are
at the same exact position for all three compounds and may be attributed to d-f
and/or f-f transitions on the uranium center. This is not unusual and has been

observed for several other compounds with U4+ ions.”l6

Another explanation
may be that that water is being absorbed onto the crystals and some of these peaks

(3511 and 1613 cm’l) are coming fiom the O-H bending modes of water.

101

V .
____._ m. w- v-

 

0.7

   

(A)

    
 

% Reflectance .
O
U!

5600 4800 4000 3200 2400 1600 800
Wavenumber (cm‘l)

IrrrIIjjjII|

1.2 ‘

(B)

   

1.1

 
 
  

1.0
0.9

O 0 00.0..

0.8

% Reflectance

0.7

I I L l I I. I l l-‘

LJ ILJ I L l LJJJJJ Ll I

5600 4800 4000 3200 2400 1600 800
Wavenumber (cm’)

0.6 7

 

0.8 ~ ‘ r11 rri‘ I1 Ij Irf‘fi I ITI—l—I fi' I

(C)

% Reflectance

 

I l l I l 1 A

JLJIIL-lllLkLlJLLll

5600 4800 4000 3200 2400 1600 800
Wavenumber (cm")

0.5

Figure 3.8 Diffuse reflectance optical spectra for (A) CSCUUTC3, (B)
RbCuUT‘e; and (C) KCuUTe; (in the Mid-IR region)

102

Cation Efiect on ACuUT e 3 - For the purpose of this study, attempts were
also made to synthesize the isostructural NaCuUTe3 and LiCuUTe3 members. The
x-ray powder x-ray diffraction patterns of these products, however, indicated that
the [type a] structure was not formed. For the sodium synthesis, the compound
formed was indeed NaCuUTe3 but it now adopts another structure type of the
AMM’Q; family [type b] (see Figure 3.1). This is not surprising since many
members of this [type b] structure type contain sodium as its cation (e.g.;
NaCuTiS3, NaCquSe3, and NaCquTe3). This change in structure is due solely to
the fact that a different alkali metal was used. In both structure types [type a and
b], the basic building blocks are [MQ4] tetrahedrons (TET) and [M’Qg]
octahedrons (OCT). The difference lies in the way that they order across the layer.
In the first [type a] structure type, the repeat pattern across the layer is ~TET —
OCT - TET -- OCT~. In the second [type b] structure type, the repeat pattern
changes to ~TET - TET - OCT — OCT~. Each building block is now paired and
these pairs alternate across the layer, consequently causing the coordination
environment around the alkali metal to be reduced from bicapped trigonal
prismatic [type a] to monocapped trigonal prismatic [type b]. Because this
reduced coordination environment is more stable for a sodium cation than any
larger cation, this [type b] structure type forms.

For the lithium synthesis, the x-ray powder diffraction pattern was less
conclusive. While it bears some resemblance to that of the pattern for the second

structure type [type b], there are many more peaks that could not be indexed. One

103

could postulate that since lithium is even smaller, the structure could have
undergone another modification where the layers have now come so close to one
another that they have joined together to form a three-dimensional framework.
This is called the “counterion effect”.17 The third and fourth structure types
[types c and d] that make up the AMM’Q family are three-dimensional and it is
therefore feasible that the lithium compound could adopt one of these structures.
But upon carefill comparison of the x-ray diffraction patterns, it was determined
that this is not the case. Another possibility is that the lithium did not incorporate
into the product and a ternary CunyTeZ phase has formed. If this were true, the
excess LizTex would have washed away during isolation in DMF as a flux.
However, no color change was observed in the DMF solution during this step. To
really determine which compound has formed, crystals of suitable size are needed

for single crystal x-ray diffiaction studies and this has not been done.

104

 

.‘L‘t

.ng
[\H

,.
111:

‘6 Q,“
tut.

D. Conclusions

The structure [type a] of AMM’Q3 is very simple. Made up of two very
basic building blocks, [the [MQ4] tetrahedra and the [M’Qs] octahedra], the
framework has proven to be very stable and is capable of adopting a wide variety
of elemental combinations. The series ACuUTe3 (A = C8, Rb, K) adopts this
structure type. Compared to KCuUSe3, the electrical properties of ACuUTe3 (A =
Cs, Rb, K) appear to be significantly enhanced. However, amongst the three
members of ACuUTe3, there is not much difference in the conductivity values.
This is most likely due to the fact that the structure is very anisotropic and the
electrical conductivity is mostly due to the carriers that are traveling parallel to
and not perpendicular to the layers. Therefore, it can be concluded that the
electrical properties of these materials are more strongly affected by the
chalcogenide chosen for the framework than the alkali metal residing between the
layers. Finally, upon investigation into this structure type [type a] with sodium, it
was determined that for NaCuUTe3 the structure changes to another member of the

AMM’Q, structure types [type b].

105

References

 

10

ll

12

l3

14

15

Mansuetto, M.F.; Kean, P.M.; Ibers, J.A. J. Solid State Chem. 1993, 105,
580.

Pell, M.A.; Ibers, J.A. Chem. Ber./Recueil 1997, 130, 1.

Sutorik, A.C.; Albritton-Thomas, J.; Hogan, T.; Kannewurf, C.R.;
Kanatzidis, M.G. Chem. Mater. 1996, 8, 751.

Mansuetto, M.F.; Keane, P.M.; Ibers, J .A. J. Solid State Chem. 1992, 101,
257.

Cody, J.A. Ibers, J .A. Inorg. Chem. 1995, 34, 3165.

Wu, P.; Christuk, A.E.; Ibers, J .A. .1 Solid State Chem. 1994, 110, 337.
Huang, F .Q.; Choe, W.; Lee, S.; Chu, J.S. Chem. Mater. 1998, 10, 1320.
Yang, Y.; Ibers, J .A. J. Solid State Chem. 1999, 147, 366.

M. Evain, U-fit: “A cell parameter refinement program”, Institut des
Matériaux, Nante, France.

(a) Wendlandt, W.W.; Hecht, H.G. Reflectance Spectroscopy; Interscience
Publishers: New York, 1966. (b) Kotiim, G. Reflectae Spectroscopy;
Springer-Verlag: New York, 1969. (c) Tandon, S.P.; Gupta, J .P. Phys.
Status Solidi 1970, 38, 363.

Matkovic, P.; Schubert, K. .1 Less-Common Met. 1977, 52, 217.

Greenwood, N. N.; Eamshaw, A. Chemistry of the Elements; Pergamon
Press: New York, 1984; p 1443.

Noel, H.; Troc, R. J. Solid State Chem. 1979,27, 123.

Cheetham, A.K.; Day, P. “Solid State Chemistry; Techniques”. Oxford:
Clarendon, 1986.

Choi. K.-S.; Kanatzidis, M.G. Chem. Mater. 1999, 11,2613.

106

 

16

17

(a) Granvold, F.; Drowart, J.; Westrum, E.F., Jr. The Chemical
Thermodynamics of Actinide Elements and Compounds; International
Atomic Energy Agency: Vienna, 1984; pp 161-200 and references therein.
(b) Clifton, J.R.; Gruen, D.M.; Ron, A. J. Chem. Phys. 1969, 51, 224. (c)
Schoenes, J. Phys Rev. 1980, 63, 301.

Kanatzidis, M.G. Phosphorous, Sulfitr, and Silicon 1994, 93, 159.

107

Chapter 4

Novel Polytelluride Compounds Containing Distorted Nets of Tellurium

108

A. Introduction

Explorations into the chemistry of complex lanthanide chalcogenides,
particularly quaternary systems of the type A/M/Ln/Q (A = alkali metal, M =
transition metal, Ln = lanthanide metal, and Q = chalcogen), suggest that phase

stabilization is most facile when the transition metal used is a coinage metal,

namely Cu. This may be rationalized by the mobile nature of Cu+ ions, even at
low temperatures, which helps diffuse the ion over long distances and facilitates
its quick arrival at phase thermodynamic minirna. The introduction of Cu into the
synthetic chemistry of lanthanide chalcogenides has already produced several
quaternary compounds, including K2Cu2CeS4,l KCuCe2S6,"2 KCuLaZSé,2
CsCuCeZS6,2 KCuCeZSeé,2 CsCuCeS3,2 and KCuUSe32. These findings were

quickly followed by a rapid expansion in this area by independent investigators,

pmducmg the compounds BaErAgSs,3 CsCuUTes,4’6 CsTiUTec4’6

CsstsUTe3o_6,4’5’6 BaMLnQ, (Ln = La, Ce, Nd, Dy; M = Cu, Ag; Q = s, Se),7’8
KLsDyzCustesf and KMBaogDyCuysTef. We have since expanded this
chemistry to include both Cu and Ag in the telluride system. One important
difference between Texz' and sz' (Q = S, Se) ions is the greater tendency for the
former to associate through Te-Te bonding interactions because of the more
diffuse nature of its orbitals.9"°’” For example, in SmTe3,12 Te-Te interactions lead
to interesting superstructures. This characteristic can easily lead to mixed valency,

Which in turn can produce interesting physical phenomena.

109

By using molten alkali metal/polytelluride fluxes, several new quaternary
phases have been discovered with the general formula AwaLnyTez (A = alkali
metal, Ln = lanthanide metal, M = coinage metal). Reactions with cerium
produced the compounds KCuCeTe4,l3 RbCuCeTe4, NaOgAngeTe4, and
K2.5Ag4.5CeQTe9, while lanthanum presented the isostructural K2_5Ag4,5La2Te9.
Although these compounds possess new structure types, they still retain
components of the known binary rare earth tellurides (NdTe3l4 and ZrSe315-type).
Since both of these binary structure types require a metal with an oxidation state >
+3, we decided to circumvent their formation completely by investigating
reactions with a divalent lanthanide metal. Only a few lanthanides are stable as +2
ions, however, including Sm (4P), Eu (4?), Tm (41”) and Yb (41“). or these,

1 .
6 Prior to our

europium was selected first because it is one of the most stable.
work, the only reported quaternary europium chalcogenide of this type was
KCuEuZS6. '7 Our explorations with europium produced CuoggliuTez,18
KCquuTe4,'8 IsiamAgZ,glau'reg‘8 and KwAngumTa, all of which are best
described as Eu2+ compounds. The compounds reported here (with the exception
of K0_65Ag2Eu1,35Te4) possess what appears to be perfectly square Te nets. With

the aid of electron diffraction, however, we find most of them to be modulated.

110

B. Experimental Section

1. Reagents — The following reagents were used as obtained: Sodium metal,
analytical reagent, Spectrum Chemical Mfg. Corp., Gardena, CA; Potassium
metal, analytical reagent, Spectrum Chemical Mfg. Corp., Gardena, CA;
Rubidium metal, 99.5%, Alfa Aesar, Ward Hill, MA.; Copper metal, electrolytic
dust, Fisher Scientific, F airlawn, NJ; Cerium metal, < 250 mesh, Alfa Aesar, Ward
Hill, MA; Lanthanum metal, 40 mesh, Cerac, Milwaukee, WI; Europium metal
powder, 99.9%, ; < 250 mesh, Alfa Aesar, Ward Hill, MA; Europium metal chunk,
99.9%, Chinese Rare Earth Information Center, Inner Mongolia, China; Tellurium
powder, 100 mesh, 99.95% purity, Aldrich Chemical Co., Milwaukee, WI;
Tellurium shots, 99.9% pure, Noranda Advanced Materials, Saint-Laurent,
Quebec, Canada. Lithium Chloride, 99.8% pure, 20 mesh, Cerac Specialty
Inorganics, Milwaukee, WI; Potassium Chloride, crystals, JT Baker, Phillipsburg,
NJ; N, N, - Dimethylformamide (DMF) was used as obtained in analytical reagent
grade from Aldrich Chemical Co., 99.8% purity, Milwaukee, WI. The europium
metal chunk was cut into fine shavings with a hacksaw and flamed under vacuum
in a sealed silica ampoule to remove the oxide coating before being used. The
tellurium shots were ground to a fine powder before being used.

Silver Powder — A silver coin (99.9% purity) weighing 31.54g was

dissolved in 250 mL of dilute (7 .SN) nitric acid. The solution was heated to 60°C
in an acid-resistant fume hood until the silver coin was completely dissolved. The

solution was neutralized with ammonium hydroxide and the silver was reduced

111

 

 

3

I1.
I:

NUT!

d}:
him

all

0311C

I ‘t'tn.
'I‘ ‘.

 

 

with formic acid until a pH of 7-8 was reached. The resulting pale grey solid was
filtered, washed with c0pious amounts of distilled water and acetone, and dried in
vacuum overnight. The final yield was 31.095g.

Sodium T elluride, NazT e — The following procedure was modified from
that given in the literature.” 8.05g (0.35 mol) Na was sliced in an N2 filled
glovebox and combined with 21.95g (0.17 mol) Te in a 1000 mL single neck
round bottom flask. This mixture represents a slight excess of Na and slight
deficiency of Te. The flask was connected to a glass adapter with a stopcock joint
and removed fi'om the glovebox. The flask and adapter was then connected to a
condenser apparatus and chilled to -78°C using a dry ice/acetone bath.
Approximately 800mL of NH3 were condensed, under an N; atmosphere, onto the
reagents, giving a dark blue solution. The solution was stirred via a Teflon coated
magnetic stir bar and the reaction mixture was maintained at -78° for up to 24
hours. The dry ice was then removed and the NH; was allowed to evaporate off as
the flask warmed up to room temperature under a constant flow of N2
(approximately 10 hours). A second portion of NH3 was added and the process
was repeated to ensure complete reaction of the reagents. The resulting peach
colored powder was evacuated on a Schlenck line for approximately 5 hours and
taken into an N2 filled glovebox where it was ground to a fine powder. Due to its
prepensity to decompose even under an inert glovebox atmosphere, the material

was stored in a glass ampoule clamped shut with a ground glass lid.

112

Potassium T elluride, KzT e — Synthesis of this material was performed as
described in Chapter 2, Section B. l.
Rubidium T elluride, Rb zTe ~— Synthesis of this material was performed as

described in Chapter 2, Section B].

2. Synthesis — All manipulations were carried out under a dry nitrogen
atmosphere in a Vacuum Atmospheres Dri-Iab glovebox.

KCuCeTe4 (I) — Amounts of 0.206g KzTe (1.0 mmol), 0.032g Cu (1.0
mmol), 0.070g Ce (0.5 mmol), 0.303g Te (3 mmol) were weighed into a vial and
mixed with 0.5g of a LiCl/KCl (45:55) eutectic flux in a N2 filled glovebox. The
reagents were then loaded into a 13 mm silica ampoule. The ampoule was
removed from the glovebox, evacuated on a Schlenck line to less than 2.0 x 104
mbar, and flame sealed. The reactants were heated to 400°C in 12 hours,
isothermed at that temperature for 10 hours, raised to 700°C in 22 hours, and
isothermed at that temperature for 5 days. The tube was then cooled to 400°C at —
3°C/hr and quenched to room temperature in 4 hours. The excess KzTex flux was
removed with successive portions of DMF, under N2 atmosphere, until the
solution remained clear. The LiCl/KCl eutectic flux was then removed by
washing with water to reveal very thin, square copper-colored plates. A small
portion of red-brown CeTe; powder was also present, but could be removed from
the plates by simply sonicating the sample. Typical yields were 46%, based on

Ce. Phase homogeneity was confirmed by comparing the powder X-ray

113

COT.

1117.1

1'31

..._‘__.* rur—fi..*.

diffraction pattern of the product against one calculated using the
crystallographically determined atomic coordinates, see Table 4.1.
Semiquantitative microprobe analysis on single crystals gave an average
composition of K1 ,oCuyMCeHgTeygg.

RbCuCeTe4 (II) — Amounts of 0.358g szTe (1.2 mmol), 0.019g Cu (0.3
mmol), 0.042g Ce (0.3 mmol), 0.612g Te (4.8 mmol) were weighed into a vial in a
N2 filled glovebox. The reagents were then loaded into a 13 mm silica ampoule.
The ampoule was removed fi'om the glovebox, evacuated on a Schlenck line to
less than 2.0 x 10‘4 mbar, and flame sealed. The reactants were heated to 400°C in
12 hours, isothermed at that temperature for 6 hours, raised to 800°C in 18 hours,
and isothermed at that temperature for 4 days. The tube was then cooled to 400°C
at -4°C/hr and quenched to room temperature in 4 hours. The excess szTex flux
was removed with successive portions of DMF, under N2 atmosphere, until the
solution remained clear. The final product consisted of very thin, square copper-
colored plates. A small portion of red-brown CeTe3 powder was also present, but
could be removed fi'om the plates by simply sonicatin g the sample.
Semiquantitative microprobe analysis on single crystals gave an average
composition of Rbo_96CuU 5Ce1_oTe3,67.

NaogAguCeTe4 (III) — Amounts of 0.208g NazTe (1.20 mmol), 0.065g
Ag (0.60 mmol), 0.042g Ce (0.30 mmol), 0.612g Te (4.80 mmol) were weighed
into a vial and loaded into a 13 mm silica ammule. The ampoule was removed

from the glovebox, evacuated on a Schlenck line to less than 2.0 x 10‘1 mbar, and

114

flame sealed. The reactants were heated to 400°C in 12 hours, isothermed at that
temperature for 12 hours, raised to 850°C in 22 hours, and isothermed at that
temperature for 6 days. The tube was then cooled to 400°C at —4.5°C/hr and
quenched to room temperature in 4 hours. The excess NazTex flux was removed,
under N2 atmosphere, with DMF to reveal very thin, square copper-colored plates.
A small portion of red-brown CeTe; powder was also present, but could be
removed from the plates by simply sonicating the sample. Typical yields were
34%, based on Ce. Phase homogeneity was confirmed by comparing the powder
X-ray diffraction pattern of the product against one calculated using the
crystallographically determined atomic coordinates, see Table 4.2.
Semiquantitative microprobe analysis on single crystals gave an average
composition of NaarAgogCeLoTego.

Kg5Ag4.5Ce2Te9 (ITO —- Amounts of 0.247g KzTe (1.20 mmol), 0.146g Ag
(1.35 mmol), 0.084g Ce (0.60 mmol), 0.612g Te (4.80 mmol) were weighed into a
vial and loaded into a 13 mm silica ampoule. The ampoule was removed fiom the
glovebox, evacuated on a Schlenck line to less than 2.0 x 104 mbar, and flame
sealed. The reactants were heated to 400°C in 12 hours, isothermed at that
temperature for 12 hours, raised to 850°C in 22 hours, and isothermed at that
temperature for 6 days. The tube was then cooled to 400°C at —4.5°C/hr and
quenched to room temperature in 4 hours. The excess KzTex flux was removed,
under N2 atmosphere, with DMF to reveal very thin, square copper-colored plates.

A small portion of red-brown CeTe3 powder was also present, but could be

115

 

removed fi'om the plates by simply sonicating the sample. Typical yields were
74%, based on Ag. Phase homogeneity was confirmed by comparing the powder
X-ray diffiaction pattern of the product against one calculated using the
crystallographically determined atomic coordinates, see Table 4.3.
Semiquantitative microprobe analysis on single crystals gave an average
composition of K2,5Ag5,3Ce2,oTe9,9.

K2_5Ag4,5La2Teg (IO —- Amounts of 0.247g K2Te (1.20 mmol), 0.146g Ag
(1.35 mmol), 0.083g La (0.60 mmol), 0.612g Te (4.80 mmol) were weighed into a
vial and mixed with 0.5g of a LiCl/KC] (45:55) eutectic flux in a N2 filled
glovebox. The reagents were then loaded into a 13 mm silica ampoule. The
ampoule was removed from the glovebox, evacuated on a Schlenck line to less
than 2.0 x 104 mbar, and flame sealed. The reactants were heated to 400°C in 8
hours, isothermed at that temperature for 10 hours, raised to 800°C in 8 hours, and
isothermed at that temperature for 5 days. The tube was then cooled to 400°C at -
4°C/hr and quenched to room temperature in 4 hours. The excess KzTex flux was
removed with successive portions of DMF, under N2 atmosphere, until the
solution remained clear. The LiC l/KCl eutectic flux was then removed by
washing with water to reveal very thin, square copper-colored plates. A small
portion of red-brown LaTe3 powder was also present, but could be removed from
the plates by simply sonicating the sample. Typical yields were 80%, based on
Ag. Phase homogeneity was confirmed by comparing the powder X-ray

diffraction pattern of the product against one calculated using the

116

crystallographically determined atomic coordinates, see Table 4.4.
Semiquantitative microprobe analysis on single crystals gave an average
composition of KgsAgsaLamTe 10, 5.
CurmEuTez (VI) — Initial investigations into the quaternary
szTe/Cu/Eu/Te system led to the serendipitous discovery of CUObéEUTCZ. The
compound was found as a minor product from a reaction mixture of 0.298 g szTe
(1.0 mmol), 0.032g Cu (0.5 mmol), 0.038g Eu (0.25 mmol), and 0.510g Te (4.0
mmol). The mixture was weighed into a vial and loaded into a 13 mm silica
ampoule. The ampoule was then removed from the glovebox, evacuated on a
Schlenck line to less than 2.0 x 10“ mbar, and flame sealed. The reactants were
heated to 400°C in 12 hours, isothermed at that temperature for 12 hours, raised to
850°C in 22 hours, and isothermed at that temperature for 6 days. The tube was
then cooled to 400°C at —4°C/hr and quenched to room temperature in 4 hours.
The excess szTex flux was removed, under N2 atmosphere, with DMF to reveal
very thin, square copper-colored plates. Phase homogeneity was confirmed by
comparing the powder X-ray diffiaction pattern of the copper-color plates in the
product against one calculated using the crystallographically determined atomic
coordinates, see Table 4.5. Semiquantitative microprobe analysis on the single
crystal used for X—ray data collection gave an average composition of
cu0.156Etll.oTt‘a.o-
After the initial discovery of Cuo,“EuTe2, fiirther reactions were carried out

to synthesize this compound in a logical fashion. Although this compound has not

117

been synthesized pure, it has been reproduced in powder form from a direction
combination reaction of the elements. This reaction consisted of a mixture of
0.042g Cu (0.66 mmol), 0.280g EuTe (1.0 mmol), and 0.128g Te (1.0 mmol) that
was heated to 800°C in 30 hours, isothermed at that temperature for 6 days, and
slow cooled to 100°C in 150 hours. The tube was then quenched to room
temperature in 4 hours. Since there was no flux used and the reaction was that of a
direct combination, no isolation step was necessary. The powder X-ray diffraction
pattern of the product clearly showed the presence of this compound, but was not
100% pure. Since the product was that of a heterogenous powder, it was not
possible to carry out firrther characterization of this material.

KCquuTe4 (VII) — Amounts of 0.l65g KzTe (0.8 mmol), 0.025g Cu (0.4
mmol), 0.061 g Eu (0.4 mmol), 0.306g Te (2.8 mmol) were weighed into a vial and
loaded into a 13 mm silica ampoule. The ampoule was removed from the
glovebox, evacuated on a Schlenck line to less than 2.0 x 104 mbar, and flame
sealed. The reactants were heated to 400°C in 12 hours, isothermed at that
temperature for 12 hours, raised to 850°C in 22 hours, and isothermed at that
temperature for 6 days. The tube was then cooled to 400°C at —4°C/hr and
quenched to room temperature in 4 hours. The excess KzTex flux was removed,
under N2 atmosphere, with DMF to reveal very thin, square copper-colored plates.
Phase homogeneity was confirmed by comparing the powder X-ray diffraction

pattern of the product against one calculated using the crystallographically

118

determined atomic coordinates, see Table 4.6. Semiquantitative microprobe
analysis on single crystals gave an average composition of KngCuroEuLoTe“.

Na0_2Agz.gEuTe4 (VIII) — Amounts of 0.208g NazTe (1.2 mmol), 0.097g
Cu (0.9 mmol), 0.046g Eu (0.3 mmol), 0.612g Te (4.8 mmol) were weighed into a
vial and loaded into a 13 mm silica ampoule. The ampoule was removed from the
glovebox, evacuated on a Schlenck line to less than 2.0 x 10‘4 mbar, and flame
sealed. The reactants were heated to 400°C in 12 hours, isothermed at that
temperature for 12 hours, raised to 850°C in 22 hours, and isothermed at that
temperature for 6 days. The tube was then cooled to 400°C at —4°C/hr and
quenched to room temperature in 4 hours. The excess NazTex flux was removed,
under N2 atmosphere, with DMF to reveal very thin, square copper-colored plates.
Typical yields were 67%, based on Ag. Phase homogeneity was confirmed by
comparing the powder X-ray diffraction pattern of the product against one
calculated using the crystallographically determined atomic coordinates, see Table
4.7. Semiquantitative microprobe analysis on single crystals gave an average
composition of NaasAgIJEuLoTeyz.

[(0,65Ag2Eu1, 35Te4 (1A9 — Amounts of 0.206g KzTe (1.0 mmol), 0.140g
EuTe (0.5 mmol), 0.054g Ag (0.5 mmol), 0.510g Te (4.0 mmol) were weighed
into a vial and loaded into a 9 mm silica ampoule. The ampoule was removed
from the glovebox, evacuated on a Schlenck line to less than 2.0 x 10‘1 mbar, and
flame sealed. The reactants were heated to 450°C in 12 hours, isothermed at that

temperature for 3 days, cooled to 150°C at —4°C/hr, and quenched to room

119

temperature in 4 hours. The excess KzTex flux was removed, under N2
atmosphere, with DMF to reveal square silver-black plates. Typical yields were
95%, based on Ag. Phase homogeneity was confirmed by comparing the powder
X-ray diffraction pattern of the product against one calculated using the
crystallographically determined atomic coordinates, see Table 4.8.
Semiquantitative microprobe analysis on single crystals gave an average

COHTPOSlthIl 0f K079Ag2,oEu1,2Te4_4,

120

Table 4.1 Calculated and Observed X-ray Powder Diffraction Pattern for
KCuCeTe4 (I)

 

 

 

h kl duic (A) dLAA) g” (obs) (%)
0 0 1 21.3040 21.0782 9.23
0 0 2 10.6520 10.6081 4.62
0 0 3 7.1013 7.0762 47.43
0 0 4 5.3260 5.3109 12.97
0 0 5 4.2608 4.2508 9.28
0 0 6 3.5507 3.5430 76.09
0 0 7 3.0434 3.0370 100.00
0 0 8 2.6630 2.6579 10.36
0 0 9 2.3671 2.3626 26.67
1 17 2.1859 2.1717 3.05
0010 2.1304 2.1266 6.18
0 0 11 1.9367 1.9331 12.31
2 14 1.8601 1.8616 4.26
0 0 12 1.7753 1.7722 1.78
0 0 13 1.6388 1.6363 1.96

 

121

Table 4.2 Calculated and Observed X-ray Powder Diffraction Pattern for
N30.8A81.2C6Te4 (HI)

 

 

h kl dQA) ding) Mobs) (%)
0 0 2 10.1255 10.1255 2.30

0 0 3 6.7503 6.7528 21.70
0 0 4 5.0627 5.0706 8.61

0 0 6 3.3752 3.3803 73.31

1 0 4 3.3424 3.3256 11.96
1 1 1 3.1103 3.0899 16.93
0 0 7 2.8930 2.8981 100.00
0 0 8 2.5314 2.5392 17.54
0 0 9 2.2501 2.2547 8.79

1 17 2.1300 2.1518 9.52

0 0 10 2.0251 2.0298 8.69

0 0 11 1.8410 1.8454 39.54
0 0 12 1.6876 1.6916 4.49

 

122

Tabll

Table 4.3 Calculated and Observed X-ray Powder Diffraction Pattern for
KngggsCezTeg (IV) (based on superstructure)

 

 

 

 

h kl rig], (A) hot) Illa. (obs) M)
0 4 0 12.6103 13.5574 59.66
0 6 0 8.4069 8.5382 100.00
0 8 0 6.3052 6.4468 14.92
0 10 0 5.0441 5.1244 14.12
0 14 0 3.6030 3.6338 6.01
0 16 0 3.1526 3.1804 23.37
2 12 -1 2.9792 2.9852 9.92
0 18 0 2.8023 2.8254 8.55
3 15 0 2.6872 2.6996 15.45
0 20 0 2.5221 2.5346 0.97
3 17 0 2.4721 2.4880 2.69
4 14 -1 2.3731 2.3899 8.90
0 22 0 2.2966 2.2485 9.41
0240 2.1017 2.1217 7.28
0 26 0 1.9401 1.9563 4.13
3 25 0 1.8390 1.8384 4.97
2 26 -1 1.7629 1.7628 2.38
2 10 2 1.7397 1.7328 2.41
3 7 2 1.6656 1.6656 1.59
2 30 0 1.6309 1.6318 3.28
8 10 0 1.5907 1.5890 2.62
0 32 0 1.5763 1.5756 2.42

 

‘

123

Table 4.4 Calculated and Observed X-ray Powder Diffi'action Pattern for
KZSAgtsLtuTeg (V) (based on superstructure)

 

 

h kl die (A) dflA) I/ILax (obs) (%)
0 4 0 12.6820 13.3791 59.7
0 6 0 8.4546 8.6594 32.7
0 8 0 6.3410 6.4301 16.1
0 10 0 5.0728 5.1072 5.2
3 7 0 3.8444 3.8414 6.3
0 l4 0 3.6234 3.6416 35.7
3 9 0 3.5334 3.5320 9.9
016 0 3.1705 3.1747 100.0
012 1 3.0100 3.0885 9.6
2 6 1 2.9923 2.9914 49.9
3 11 —1 2.9168 2.9204 27.2
4 10 0 2.8251 2.8245 45.7
2 10 1 2.7061 2.7050 70.2
0 20 0 2.5364 2.5395 8.9
3 17 0 2.4928 2.4927 8.6
2 14 1 2.3983 2.3968 47.3
6 2 -1 2.3223 2.3278 9.1
4 16 —1 2.2531 2.2550 47.9
5 15 0 2.1200 2.1231 67.3
0 26 0 1.9511 1.9576 44.2
4 14 —2 1.8457 1.8476 4.2
0280 1.8117 1.8189 2-7

 

 

 

 

x

124

 

11111

Pine

Table 4.4 continued Calculated and Observed X-ray Powder Diffraction
Pattern for KgsAgagLazTe9 (V) (based on superstructure)

 

 

h kl MA) M) Mobs) (%)
6 10 -2 1.7652 1.7658 4.8
515 1 1.7379 1.7375 19.8
8 2 0 1.6969 1.6967 21.7
6 14 -2 1.6707 1.6713 12.6
0 20 2 1.6372 1.6351 13.3
4 2 2 1.5965 1.5969 41.8
3 15 2 1.5454 1.5456 21.7

 

125

 

 

 

126

Table 4.5 Calculated and Observed X-ray Powder Diffiaction Pattern for
CumggEuTe2 (VI)

1 kl «Lane and) we) (’4)

0 0 3 3.4201 3.4879 16.40

1 0 2 3.3749 3.3639 17.41

1 l 1 3.0275 3.0004 10.62

1 l 2 2.6958 2.6645 100.00

1 l 3 2.3243 2.3222 28.74

2 0 0 2.2405 2.2951 11.66

0 1 4 2.2261 2.2151 24.29

1 2 1 1.9668 1.9641 11.40

0 1 5 ‘ 1.8657 1.8506 8.76

006 1.7100 1.7137 10.19

0 2 4 1.6874 1.6656 16.19

0 l 6 1.5977 1.6039 2.69

2 2 0 1.5792 1.5737 19.64

 

Tab
110

Table 4.6 Calculated and Observed X-ray Powder Diffraction Pattern for

KCquuTe4 (VII)

 

 

b kl afloat) 6&4) ML” (obs) (%)
001 11.3650 11.3156 15.41
0 0 2 5.6825 5.6657 30.34
0 0 3 3.7883 3.7750 100.00
1 10 3.1371 3.1213 6.02
0 0 4 2.8413 2.8326 54.94
0 0 5 2.2730 2.2669 14.12
0 0 6 1.8942 1.8897 7.53
00 7 1.6215 1.6169 7.07

 

 

 

127

Table 4.7 Calculated and Observed X-ray Powder Diffraction Pattern for
N30.2Ag2.8EUTC4 (VIII)

 

 

b kl a“;E (A) dLbéA) 104w (obs) cm
0 0 1 11.1120 11.2785 28.75
0 0 2 5.5560 5.5304 44.89
0 0 3 3.7040 3.6898 27.27
1 0 2 3.4767 3.4559 18.75
1 0 3 2.8488 2.8506 11.59
0 0 4 2.7780 2.7889 76.59
1 12 2.7414 2.7477 100.00
1 1 3 2.4004 2.4058 11.25
1 0 4 2.3576 2.3592 17.27
2 0 0 2.2287 2.2357 21.14
1 14 2.0840 2.0922 27.73
21 0 1.9934 1.9934 17.73
2 0 3 1.9096 1.9050 11.14
0 0 6 1.8520 1.8585 24.89
2 0 4 1.7384 1.7450 1.011
1 0 6 1.7102 1.6820 8.41
2 14 1.6196 1.6384 8.75
1 16 1.5967 1.5987 8.64
2 2 0 1.5759 1.5753 9.55

 

128

 

Tab

Table 4.8

K0.65A82EUI.35TC4 (IX)

Calculated and Observed X-ray Powder Diffraction Pattern for

 

 

 

h kl dag!» 6&6!» [IL-Lu (obs) (%)
0 0 2 11.3997 11.3421 8.22
0 0 4 5.6999 5.6901 43.79
0 0 6 3.7999 3.7866 42.06
1 0 4 3.5314 3.5358 2.39
0 14 0 3.2342 3.2333 12.36
0 0 8 2.8499 2.8426 110.00
1 10 4 2.7846 2.7831 7.97
1 4 8 2.3549 2.3514 3.69
1 171 2.2804 2.2892 1.94
0 20 0 2.2640 2.2622 2.84
2 0 0 2.2495 2.2508 2.18
0 20 2 2.2206 2.2294 2.40

1 10 8 2.1257 2.1223 0.96
0 22 0 2.0891 2.0873 1.20

1 16 6 2.0264 2.0274 1.04
2 5 5 1.9691 1.9771 0.69
2 10 4 1.8994 1.8957 26.29
2 13 3 1.8338 1.8345 1.64
0 20 8 1.7727 1.7713 1.33
2 13 7 1.6345 1.6341 2.75
13 13 1.6245 1.6253 4.64
2 20 0 1.5957 1.5952 0.73

 

129

3. Physical Measurements - The instrumentation and experimental setup
for the following measurements are the same as described in Chapter 2, Section
B.3: Semiquantitative Energy Dispersive Spectroscopy (EDS), Powder X-ray
Diffraction (PXRD), Transmission Electron Microscopy (TEM), Infrared
Spectroscopy (IR), Magnetic Susceptibility Measurements, and Charge Transport
Measurements.

Single Crystal X -ray Difiiaction — Intensity data for KCuCeTe4 (I),
NaugAngCeTe.‘ (HI), and KCquuTe4 (VII) were collected on a Rigaku AFC6S
four-circle automated difii’actometer equipped with a graphite crystal
monochromator. A single crystal of each was mounted on the tip of a glass fiber.
The unit cell parameters were determined from a least-squares refinement using
the setting angles of 20 carefully centered reflections in the 8° 5 20 3 30° range.

The data were collected with an 00-20 scan technique over one-half (I), one-quarter

(III), and the fill (VIII) sphere of reciprocal space, up to 60° in 20. Crystal
stability was monitored with three standard reflections whose intensities were
checked every 150 reflections. No significant decay was detected during the data
collection periods. The data was corrected for Lorentz-polarization effects and an
empirical absorption correction based on ‘P-scans was applied to all data during
initial stages of refinement. The structures were solved by direct methods using

the SHELXS-8620 package of crystallographic programs and full matrix least

squares refinement was performed using the TEXSAN sofiware package”.

130

Complete data collection parameters and details of all structure solutions and

refinements are given in Tables 4.19 and 4.11.

Intensity data for RbCuCeTe4 (II), szAgrsCezTeo (IV), K2.5Ag4.5La2Te9
(V), CUoooEllTCz (VI), Nao.2Agz.sEuT94 (VIII), and K0.65Ag2Eu1.35Te4 (IX) were
collected on a Siemens SMART Platform CCD diffractometer using graphite
monochromatized Mo Ka radiation. A single crystal of each was mounted on the
tip of a glass fiber. The data were collected over one-half (II, VI) and a full (1V, V,
VIII, IX) sphere of reciprocal space, up to 50° in 20. The individual fi’ames were
measured with an a) rotation of 03° and acquisition times ranging fiom 40 to 80
sec/frame. The SMART22 software was used for the data acquisition and
SAINT23 for the data extraction and reduction. The absorption correction was
performed using SADABS.24 The structure was solved by direct methods using
the SHELXTL” package of crystallographic programs. Complete data collection

parameters and details of all structure solutions and refinements are given in

Tables 4.9-4.11.

131

Table 4.9 Crystallographic Data for KCuCeTe4 (I), RbCuCeTea (II), and

NaosAngCeT64 (III)

 

 

KCuCeTe4 RbCuCeTe. N a. ,sAg. JCCTC4
crystal habit, color plate, copper plate, copper plate, COpper
Difl'ractometer Rigaku AF C6S Siemens SMART Rigaku AF C68

Platform CCD
Radiation Mo-Ka (0.71069A) Mo-Kor (0.71073A) Mo-Ka (0.7106914)
Crystal Size, mm3 0.29 x 0.29 x 0.03 0.18 x 0.31 x 0.02 0.23 x 0.23 x 0.02
Temperature, K 293 173 293
Crystal System orthorhombic orthorhombic orthorhombic
Space Group Pmmn (#59) Pmmn (#59) Pmmn (#59)
3, 4.436(1) 4.4330(9) 4.450(3)
b, A 4.4499(9) 4.4697(9) 4.448(4)
c, A 21.304(2) 21 .951(4) 20.25(1)
v, A3 420.5(4) 434.9505) 401.3(5)
Z 2 2 2
n, mm" 22.020 26.219 22.690
indexranges 05h55 -55h50 05h55

-55k55 -55k55 05k55

~2551525 ~285l528 -245l524
29m, deg 50 50 50
sec/flame N/A 40 N/A
total data 1701 2645 1837
unique data 498 641 815
Ram) 0.065 0.156 0.032
no. parameters 30 3O 28
final R/Rw‘, % 6.36/5.76 N/A 7.20/9.20
finalRl/wRZb, % N/A 1034/2737 N/A
GOP 2.609 N/A 4.92
Goof N/A 1.048 N/A

 

784617,! - fiiJ)/f[1~‘,l RW={Z[W(F0'FC)2]/Z[W(F0)2]}m
bm=2(|r,|.l1:.|)/z|1:,| wR2={Z[W(Foz-Fc2)2]/Z[W(Foz)2]}m

132

Table 4.10

Crystallographic Data for K2,5Ag4_5Ce2Te9 (IV), K2,5Ag4,5La2Te9

 

 

(V), and CUoraEuTez (VI)

KuAgucezTeg KuAgMLazTe, Cup,“EuTe2
crystal habit, color plate, copper plate, copper plate, copper
Diffi‘actometer Siemens SMART Siemens SMART Rigaku AFC6S

Platform CCD Platform CCD
Radiation Mo-Kot (0.71069A) Mo-Ka (0.71073A) Mo-Ka (0.71069A)
Crystal Size, mm3 0.31 x 0.22 x 0.04 0.44 x 0.13 x 0.02 0.44 x 0.13 x 0.02
Temperature, K 293 293 293
Crystal System Orthorhombic Orthorhombic Tetragonal
Space Group Immm (#71) [mm (#71) P4/mmm (#129)
a, A 4.4844(9) 4.5224(6) 4.4810(16)

b, A 4.5116(9) 4.51 10(9) 4.4810(16)
c, A 50.85900) 50.61800) 10.260(3)
V, A3 1029.0(4) 1032.6(4) 206.0202)
2 2 2 2

11mm" 21.514 21.166 32.196
indexranges ~55h55 -55h55 O5h55

-55k54 -65k56 -55k55

-655l564 -655l563 -1151512
29m, deg 50 50 50
sec/flame 6O 80 N/A
total data 3348 3271 1169
unique data 754 777 141
Rant) 0.0761 0.1186 0.1083
no. parameters 41 41 61
finalRl/wRZ‘, % 821/2240 705/1703 794/2507
GooF 1.129 1.008 1.477

 

‘Rl=2(lF.l-IF.I)/zlr.l sz={lemz-FXYI/ZIMFoZYIV”

133

Table 4.11 Crystallographic Data for KCquuTe4 (VII), NaopAnguTea (VIII),
and K0.65Ag2Eul.35Te4 (IX)

 

 

KC|lelch4 NamAguEuTea K055Ag2EIIL35Te4
crystal habit, color plate, copper plate, copper plate, copper
Difli‘actometer Rigaku AF C6S Siemens SMART Siemens SMART

Platform CCD Platform CCD
Radiation Mo-Ka (0.71073A) Mo-Ka (0.71073A) Mo-Ka (0.71073A)
Crystal Size, mm3 0.40 x 0.40 x 0.04 0.l3x 0.39 x 0.01 0.21 x 0.21 x 0.02
Temperature, K 293 293 293
Crystal System Tetragonal Tetragonal Orthorhombic
Space Group P4mm (#99) P4mm (#99) Abm2 (#39)
a, A 4.4365(6) 4.4544(6) 4.4989(9)
b, A 4.4365(6) 4.4544(6) 45.279(9)
c, A 1 1.365(2) 11.106(2) 22.799(5)
v, A3 223.69(6) 220.37(6) 4644.406)
Z 1 1 20
u. mm" 24.789 26.044 25.682
indexranges -55h55 -5_<.h55 -55h55

-55k55 -55k55 -575h557

-135l5l3 -135l514 -295h529
20...“, deg 50 50 50
sec/fi'ame N/A 40 80
total data 1597 1941 22152
unique data 295 361 5116
R(int) 0.1473 0.1 1 14 0.0824
no. parameters 23 23 180
final R1/wR2', % 7.35/ 17.88 730/2035 646/2780
Goof 1.198 1.078 2.148

 

“Mr-2012.! JFJszFJ w82={lelFoz-FJYI/ZIwIFffli‘”

134

C. Results and Discussion
1. A,M(z-,)CeTe4 (I -III)
Structure Description of AxM(2.x)Ce T e4 (A = Na, K, Rb; M = Cu,Ag) (I -III)
KCuCeTe4 forms a two-dimensional structure composed of two "distinct" layers,
[CuTej' and [CeTe3], see Figure 4.1. Because each layer is known to exist

independently, KCuCeTe4 can be regarded an intergrowth compound and a more

descriptive formula would be K+[CuTe]-[CeTe3]. The [CuTe]' layer can be

described as an ideal anti-PbO structure type, made up of ribbon tetrahedral

[CuTe4] units that share edges in two dimensions, see Figure 4.2A. The [CeTe3]

layer adopts the NdTe3 structure type, and has 9-coordinate Ce in a monocapped

square anti-prismatic environment of Te, see Figure 4.2B. The [CuTe]- and
[CeTe3] layers are separated by potassium ions and stack in an A-B-A-B order
parallel to the c-axis. The potassium cations are stabilized between the layers in a
square antiprismatic coordination environment of Te, see Figure 4.2C.
RbCuCeTe4 and NaugAngeTe4 are isostructural to KCuCeTe4. The non-
stoichiometry in the formula of NaagAglpCeTea is due to a 20% disorder on the
sodium site with silver. The fiactional atomic coordinates, isotropic and
anisotropic temperature factors, bond distances, and bond angles for all three
compounds is given in Tables 4.12-4.16.

The [CeTe3] layer contains a square Te lattice net, see Figure 4.2D. Within

this net, all the Te-Te distances are equal at 3.14 A, which is longer than the the

135

normal Te-Te bond of 2.8 A but shorter than the van der Waals contact of
3.8-4.0 A. Tellurium square nets are rare and only few are known, e.g.,
LnTez, Ln2Te5, and LnTe3,26 CSThzTe6,” K0.33Ba0.67AgTe2,” Cs,Te,,,28 and
ALn3Te8 (A = Cs, Rb , K; Ln = Ce , Nd)9. Depending on the electron-count,
these nets can have different electronic structures, which can lead to
instabilities and structural distortions within them.29 These distortions can
lead to interesting physical phenomena such as charge density waves and
anomalies in the charge transport properties. When the formal oxidation
state of all Te atoms in the net is -2, a stable square net is observed (e. g.
NaCuTe).3O However, when the formal oxidation state is < -2, or when there

are atomic vacancies in the square net, structural distortions are possible
leading to Te-«Te bonding interactions and the formation of Tex“. species

within the net. These distortions are manifested through the formation of a
, superstructure with respect to the ideal square net.ll
An intriguing fact about this structure is that the [C uTe]" layer in

KCuCeTe4 is different from that found in KCuTe.31 In KCuTe, a layered

boron nitride type structure is seen while the [C uTe]" layer with the same
alkali ion in the quaternary phase adopts the structure of NaCuTe (i.e. anti-

PbO), see Figure 4.3. So, in a way, the KCuCeTe4 has enforced a higher

energy, presumably metastable, structure on KCuTe by sandwiching it

136

between CeTe3 layers. The driving force for the stability of KCuCeTe4 is

therefore unknown unless some electron transfer exists between the two

different metal chalcogenide layers.

137

 

 

 

 

Figure 4.1 ORTEP representation of the extended structure of KCuCeTe4 as
seen down the b-axis (90% probability ellipsoids). The ellipses with octant

Shading represent Ce, the crossed ellipses represent Cu, the large open ellipses
rcpresent K and Te.

138

(A) (B)
Qua

 

 

 

 

 

Figllre 4.2 ORTEP representation (90% probability ellipsoid) of (A) a view
Perpendicular to the [CuTe‘] layer of KCuCeTe4, (B) the coordination environment
around Ce in KCuCeTe4, (C) the coordination environment around K in
KCuCeTet, and (D) a view perpendicular to the [CeTe3] layer in KCuCeTe4. The
ellipses With octant shading represent Ce, the crossed ellipses represent Cu, the

large Open ellipses represent K and Te.

139

Table 4.12 Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Ueq) for KCuCeTe4 (I), RbCuCeTe4 (II), and
NaoyAngeTea (III) with Estimated Standard Deviations in Parentheses.

 

 

 

 

 

 

 

 

KCuCeTe4

atom x y z occupancy Beq“, A2
Ce 0.25 0.25 0.4022(1) 1.0 1.07(8)
Te(l) 0.25 0.25 0.5565(1) 1.0 1.10)
Te(2) 0.25 0.75 0.2827(1) 1.0 1.7(1)
Te(3) 0.75 0.25 0.2829(1) 1.0 1.40)
Te(4) 0.25 0.25 0.0703(1) 1.0 1.70)
Cu 0.25 0.75 0.9992(3) 1.0 2.6(2)
K 0.75 0.75 0.1427(4) 1.0 1.7(3)
RbCuCeTe4

atom x y z occupancy Ueqb, A2
Ce 0.25 0.25 0.4050(1) 1.0 1.00)
Te(l) 0.25 0.25 0.5548(1) 1.0 0.9(1)
Te(2) 0.25 0.75 0.2893(2) 1.0 1.8(1)
Te(3) 0.75 0.25 0.2893(2) 1.0 1.60)
Te(4) 0.25 0.25 0.0677(2) 1.0 1.8(1)
Cu 0.25 0.75 0.0001(3) 1.0 2.6(2)
E 0.75 0.75 0.1475(2) 1.0 2.0(1)
anrAgtzCeTert

atom x y z occupancy Beqaa AZ
Ce 0.25 0.25 0.3966(1) 1.0 054(3)
1‘90) 0.25 0.25 0.5594(1) 1-0 053(4)
Te(2) 0.25 0.75 0.2716(1) 1-0 103(4)
1‘90) 0.75 0.25 0.2711(1) 1-0 193(4)
Te(4) 0.25 0.25 0.0919(1) 1.0 098(4)
A80) 0.25 0.75 0.99950) 1-0 130(5)
A80) 0.75 0.75 0.1411(4) 0-2 03(1)
Na 0.75 0.75 0.1411(4) 0-3 93(1)

 

 

'3 values for anisotro ' ll fined t are iven in the form of the isotropic

pica y re a oms g 2
eqUiValent displacement parameter defined as Bar. = (4/3)[azB(1,1) + b23052) 4'. C 13%;)
+ ab(cosy)B(l,2) + ac(cos|3)B(l,3) + bc(cosa)B(2,3)]. I’Ueq 18 defined as one-thlrd o e
trace of the orthogonalized Ug- tensor.

140

Table 4.13 Anisotropic Displacement Parameters (A) for KCuCeTe4 (I),
RbCuCeTe4 (II), and Nao_3Ag]_2CCTC4 (III) with Standard Deviations in

Parentheses.

 

 

 

 

 

 

 

 

KCuCeTe4

atom U11 022 U33 U12 013 U23
Ce 0.0129(8) 0.0064(7) 0.022(1) 0 0 0
Te(l) 0.0116(9) 0.0051(8) 0.025(2) 0 0 0
Te(2) 0.028(1) 0.0104(8) 0.025(2) 0 0 0
Te(3) 0.025(1) 0.0114(8) 0.018(2) 0 0 0
Te(4) 0.019(1) 0.0111(9) 0.036(2) 0 0 0
Cu 0.037(2) 0.031(2) 0.032(4) 0 0 0

K 0.032(4) 0.032(4) 0.000(4) 0 o 0
RbCuCeTe4

atom Ull U22 U33 U12 U13 U23
Ce 0.003(1) 0.009(1) 0.018(2) 0 0 0
Tea) 0.001(1) 0.007(1) 0.019(2) 0 0 0
Te(2) 0.018(2) 0.017(2) 0.018(2) 0 0 o
Te(3) 0.014(2) 0.017(2) 0.016(2) 0 0 0
Te(4) 0.006(2) 0.019(2) 0.030(2) 0 0 0
Cu 0.015(4) 0.029(4) 0.033(4) 0 0 0
Rb 0.014(2) 0.021(2) 0.025(3) 0 0 0
Nao.tAgi.2CeTe4

atom 011 022 U33 012 013 023
Ce 0.0050(6) 0.0064(7) 0.022(1) 0 0 0
T9(1) 0.0050(7) 0.0051(8) 0.025(1) 0 0 0
Tea) 0.0149(8) 0.0104(8) 0.025(1) 0 0 0
Te(3) 0.0131(8) 0.0114(8) 0.018(1) 0 0 0
Te(4) 0.0125(9) 0.0111(9) 0.036(1) 0 0 0
&) 0.016(1) 0.031(2) 0.032(1) 0 0 0

 

141

Table 4.14 Selected Distances (A) and Bond Angles (deg) for KCuCeTe4 (I)
with Standard Deviations in Parentheses.

 

 

 

Bond Distances

Ce — Tel 3.263(1) x 4 3.287(4) x 1

Ce—Te2 3.382(3) x 2

Ce — Te3 3.374(3) x 2

Te2 -Te3 3.1417(6) x4

Cu—Te4 2.667(4) x 2 2.692(4) x 2

K - Te2 3.717(7) x 2

K — Te3 3.725(7) x 2

K — Te4 3.499(4) x 4

Bond Angles

Tel — Ce -Te1 74.35(5) x 4 85.66(4) x 2 148.7(1) x 2
Tel — Ce — Te2 7579(5) x 4 130.68(6) x 2 138.86(4) x 2
Tel — Ce — Te3 7590(5) x 3 130.56(6) x 3 138.89(4)x 2
Te2 — Ce — Te3 5543(4) x 2 8221(9) x 1 8229(9) x 1
Te2 — Te3 -— Te2 8992(2) x 2 9018(2) x 2 179.8(2) x 2
Te4—Cu—Te4 108.22(5) x2 111.5(2) x1 112.5(2) x1
Te2—K—Te2 73.3(2) x1

Te2 - K — Te3 49.9(1) x 4

Te2 — K -— Te4 88.57(7) x 4 137.0(1) x 4

Te3 —K—Te3 73.4(2) x 2

Te3 — K — Te4 88.47(7) x 4 137.( 1) x 4

1‘94 - K — Te4 79.00) x 2 78.70) x 2 127.7(3) X 2

 

 

142

 

Table 4.15 Selected Distances (A) and Bond Angles (deg) for RbCuCeTe4 (II)
with Standard Deviations in Parentheses.

 

 

 

Bond Distances

Ce-Tel 3.269(1) x 8 3.288(4) x1

Ce — Te2 3.384(3) x 2

Ce—Te3 3.371(3) x 3

Te2 - Te3 3.1476(5) x 7

Cu — Te4 2.669(5) x 3 2.683(4) x 3

Rb — Te2 3.820(5) x 2

Rb — Te3 3.832(5) x 3

Rb—Te4 3.602(3) x 7

End Angles

Tel — Ce -Te1 74.33(8) x 4 85.8(4) x 2 148.66( 15) x 2
Tel — Ce - Te2 75.60(6) x 4 l30.87(7) x 4 l38.66(4) x 1
Tel — Ce — Te3 7597(5) x 4 130.49(7) x 4 138.88(5) x 2
Te2 — Ce — Te3 5555(5) x 4

Te2 — Te3 — Te2 89.52807) x 4 90.47206) x 4 179.960 8) x 4
Te4 — Cu — Te4 107.96(6) x 4 1112.3(3) X 2

T62 - Rb — Te2 70.93(11) x 1
Te2 — Rb - Te3 48.58(6) x 4

Te2 — Rb — Te4 92.24(6) x 4 l38.86(9) X 4

Te3-Rb—Te3 71.36(11)xl

Te3 — Rb — Te4 9190(6) x 4 139.19(9) x 4

Te4 - Rb — Te4 7595(8) x 2 7669(8) x 2 121-3009) x 2

 

 

143

 

Table 4.16 Selected Distances (A) and Bond Angles (deg) for NapgAglpCeTe.)

(III) with Standard Deviations in Parentheses.

 

 

 

Bond Distances

Ce - Tel 3.272(1) x 4 3.296(3) x 1

Ce—Te2 3.371(2) x 2

Ce — Te3 3.378(3) x 2

Te2—Te3 3.148(1)x4

Agl — Te4 2.895(3) x 2 2.908(2) x 2

Na—Te2 3.455(6) x 2

Na — Te3 3.449(6) x 2

Na — Te4 3.302(3) x 4

Bond Angles

Tel — Ce -Tel 74.l9(5) x 4 85.71(5) x 2 148.74(l) x 2
Tel — Ce —— Te2 7582(4) x 4 l30.85(5) x 4 l38.66(3) x 2
Tel —Ce-Te3 7593(5) x 4 130.77(5) x4 138.790) x 4
Te2 — Ce — Te3 55.6l(4) x 4

Te3 — Ce — Te3 8242(3) x 1

Te2 - Te3 — Te2 8996(4) x 2 90.040) x 1 179.6(1) x 1
Te4—Ag] —Te4 99.9(1)x1 100.4(1)x1 114.300) x4
Te2 — Na — Te2 80.2(2) x 1

Te2 — Na — Te3 54.3(1) x 4

Te2—Na—Te4 78.270) x 4 131.7(1) x 4

Te3 -— Na — Te3 80.4(2) x l

Te3—Na—Te4 78.16(6)x4 131.7(1)x4

Te4 - Na — Te4 84.75(8) x l 8582(8) x 1 144.9(3) x 1

 

 

144

(A) NaCuTe (B) KCuTe

 

 

 

 

 

Figure 4.3 Side by side comparison of the layers in (A) NaCuTe to those in (B)

KCuTe. Both structures are viewed down the b-axis.

145

and

be}

rel

Transmission Electron Microscopy of KCuCeTe4 (I) ~— The Te net in the

[CeTe3] layer of KCuCeTe4 is fully occupied. However, the formal oxidation state
of the Te atoms in this net is —0.5, indicating the possibility of a distortion within

the Te net. Notice that the [CuTe]' layer also has a square Te net; however, the 2-
formal charge on each Te atom in this layer is not expected to lead to a distortion
and therefore a superstructure. Consequently, any observed superstructure must
be localized on the [CeTe3] part of the compound.

To probe for this distortion, electron diffraction studies were performed on
both KCuCeTe4 and Nap, 3Ag1 ,ZCeTea. Indeed, a superstructure was revealed for
both compounds along the ab plane. The supercell reflections are very weak, and
occur along both the a* and b* directions. Figure 4.4A shows an electron
diffiaction pattern of KCuCeTe4 along the ab plane. A densitometric intensity scan
obtained from the (hk0) reciprocal plane along the (lk0) row of reflections is

shown in Figure 4.43. The reflections between the (1 -1 0), (l 0 0), and (1 1 O)

reflections are due to the 0.34%“ superlattice, WhiCh corresponds to a 2-87 X bsub

(i.e.“ 13A) lattice dimension. These supercell reflections also occur along the a*
direction, at the same exact position, giving an incommensurate supercell of

“2.87am x 2. 87bsub”. However, long axial photographs taken on the x-ray

diffractometer for Na0.8Agl.2CeTe4 suggest that the incommensurate supercell
exists only along the b-axiS, giving a “laser x 2-8 7 brub” supercell. It is possible

that the “2. 87am x 2. 8 7b,,d,” supercell revealed by electron diffraction is simply a

146

twinning effect whereby two crystals with a “lamb x 2.87bwb” supercell are rotated
90° with respect to one another and superimposed under the electron beam, see
Figure 4.5. Without further evidence, however, a decision cannot be made with
certainty as to the real identity of the supercell. In either case (see Figure 4.6),

there exists a distortion within the Te net of the [CeTe3] layer, resulting in an

oligomerization of the Te atoms such that the larger unit cell redescribes the Te
net’s periodicity. Since the supercell is both weak and incommensurate, the single
crystal X-ray data needed to actually solve the superstructure could not be
obtained at Michigan State University. A collaboration has been established with
Professor Michel Evain at the Institute des Matériaux in Nante, France who
specializes in solving these types of incommensurate superstructures.
Unfortunately, attempts thus far to collect x-ray data have been unsuccessful due

to inadequate crystal quality.

147

Figure 4.4 (A) Selected area electron diffraction pattern of KCuCeTe4 with the
beam perpendicular to the layers ([001] direction) showing the 2.87asub x 2.87bsub

superlattice. (B) Densitometric intensity scan along the b* axis of the electron
diffraction pattern (boxed area in photograph) showing the (lkO) family of
ICflections. The three reflections from the sublattice of KCuCeTe4 are indexed.

The two weak peaks are from the superlattice with bsuper = 2.87bsub.

148

Electron Diffraction of KCuCeTe4

(A) .' .'

 

 

B
( ) i=3 (110) (110)
2' 0.348b*
% 0.348b*
S \ (100)
E
i

 

 

 

Reciprocal Angstroms

149

(A) o o e o

-' -' -' “1ax2.87b”

(B) o o o e

’ .. °.. ' . ' “lax2.87b”
. ' . ' . ' . rotated 90°
C . O . O . O

(C)o. o e e

“2.87ax2.87b”

a* b* 0.0 .0. 0.0
V .00....00.

lfigure 4.5 Cartoon Schematic of an electron diffraction pattern for (A) a “laser
x 2.87bmb” supercell, (B) two “1am x 2.87 M,” supercells rotated 90° with respect

to one another and superimposed under the electron beam, and (C) an apparent

“2'87asub x 2.87bm,” supercell.

150

2.87asub x 2.876%

 

Figure 4.6 View of the Te "net" in KCuCeTe4 showing (A) a“1a...t x 2.87m...”

supercell and (B) a “2.87am, x 2.87b,.,.,” supercell. 11w crystallographically
detennincd sublattice is shown in the shaded box for both.

151

Magnetic Susceptibility Measurements of AXMW ) Ce T e, (I —III) — The
magnetic susceptibilities of KCuCeTe4 RbCuCeTe4, and NaagAngeTea were
measured over the range 5-300K at 60006, see Figure 4.7. A plot of l/XM vs T for

each show that the materials exhibit nearly Curie-Weiss behavior with only slight

deviations fiom linearity beginning below 50K. Such deviation has been reported
for several Ce3+ compounds and has been attributed to crystal field splitting of the
cation's ZFS,2 ground state.32 At temperatures above 100K, um. values of 2.62 113
for KCuCeTe4, 2.78 113 for RbCuCeTea, and 2.75 [LB for NapgAngeTe4 have
been calculated. These values are in accordance with the usual range for Ce3+
compounds (2.3-2.5 113). These magnetic moments suggest that the cerium in

these compounds are trivalent with a 5fl electronic configuration. The Weiss
constants, 0, were calculated to be —3 1K for KCuCeTe4, -69K for RbCuCeTea, and
—98K for Nao_3Agl,2CeTe4. These large negative values indicate that there is a

certain amount of antiferromagnetic ordering.

152

 

 

 

 

 

 

 

 

 

 

 

 

.A .
350:-() .3
3003- . . _=
250E- . ° 5

5 E . E
if 200:- . .
150; .0 ‘5
100?- _5
50%.... .5
0’...l....1....l....l....l....

Temperature(K)

400 ...., I I I I .
350 ,(B) .3

. . :
3°05” . ° .=

: . i
250;. . :

2 : e :
150:. .0 -

E .e'

100 '. ..

5 ..o"

50?. -
0 ‘--l I I 1‘ l
0 50 100 150 200 250 300
Temperature(K)
500 n... . , I 1 I 'r‘

:(C) :
400'- .J,

. . ‘

I . .

: . . :

2300: . 1
$5 : , ° :
200- .. ° _

I ... E

100- .e'.‘ 1
O

I" -
0"“'----ln-..l....n..-.1.a.a

0 so 100 150 200 250 300

Temperature(K)

F i81111! 4.7. Inverse molar magnetic susceptibility (l/xM) plotted against
temperature (2-300K) for (A) KCuCeTe4, (B) RbCuCeTee and (C)
Na0.8Ag] , 2 CCTC4 .

153

 

KCL
(llill
Figu
indi.
how
be a
dilli
com
Na,

has;

ligl

Infrared Spectroscopy of A,M(2.,) Ce T e4 (1 —III) — The optical properties of
KCuCeTe4 (I), and NaagAngeTea (III) were determined by measuring the
diffuse-reflectance spectra of each in the Mid-IR region (6000-400 cm’l), see
Figure 4.8. The spectrum of KCuCeTe4 (I) shows no transitions in this region,
indicating that this material is metallic. The spectrum of NapgAglpCeTea (III),
however, reveals an abrupt optical transition at 0.32 eV, suggesting the material to
be a semiconductor. This difference in properties is most likely due to the
difference in the elemental makeup of the materials. However, another factor that
comes into play is the statistical disorder between the silver and sodium in
NaogAglpCeTe4 (III). It would be interesting to know how much this disorder

has an effect on the properties.

 

45

40

35

30

or/S (arbitrary units)

25

 

 

 

20 l '
00 DJ 02 03 0A 05 06 07 08

Energy (eV)

Figure 4.8 Diffuse reflectance optical Spectra 0f N30.8Agl.2CCTC4 (in the Mid'IR

region).

154

Charge Transport Measurements of A,M(2.x) Ce T at (I -—III) —— Electrical

conductivity data as a function of temperature for a hot-pressed (at 200°C) pellet
of KCuCeTe4 confirms the metallic behavior with a room temperature value of
180 S/cm (See Figure 4.9A). The thermopower data shows a Seebeck coefficient
at room temperature of ~3 uV/K, which suggests that the carriers are holes (see
Figure 4.93). The magnitude and slope of the Seebeck coefficient are consistent
with the metallic character of the sarnple. Therefore, the distortion in the Te layer,
which gives rise to the incommensurate superstructure, does not fully open up a
gap at the Fermi level of this material. This is probably due to the fact that the
distortion itself is very subtle.

The electrical conductivity data of Na0,gAg.,2CeTe4 as a function of
temperature for a room temperature pressed pellet also agree with the Optical
spectrum, showing semiconducting behavior. The data decrease with decreasing
temperature and gives a room temperature value of ~100 S/cm. The thermopower
data shows a Seebeck coefficient at room temperature of ~21 uV/K, higher than

that of KCuCeTe4.

155

 

 

 

 

 

 

 

 

 

 

 

 

250 Electrical Conductivity (A)
; ...~...~’o e j
200 L- ”.N KCuCeTe4 i
I w"?
E 150 _- 4
E5 )- I
b 100 L ...t:
50 :- NaMAguCeTe4 .1
I" l
- it
ob# 1+L14k1i IUJlnLliIJ LJLIHL
0 50 100 150 200 250 300
Temperature (K)
Thermopower (B)
L 4
20 L- ?
t NaMAguCeTe‘ '1
15 )- J
g E E
l
> 10 L
3: r 7
m '1
3 1
5 f KCuCeTe4 4
l- .... -—":
L v —- v at
o _L a n l L n n l n L n L I n n n 4 I n n . ll
50 100 150 200 250 300
Temperature (K)
Figure 4.9 (A) Four probe, electrical conductivity data of a hot-pressed pellet

of KCuCeTe4 and a room temperature pressed pellet of Nao_8Ag1_2CeTe4 as a

function of temperature.

KCuCeTe4 and a room temperature pressed pellet of NaopAglpCeTer as a

function of temperature

156

(B) Thermopower data of a hot-pressed pellet of

2.]

(on
in 11
dou

mac

neg:

310i

3qu

360‘

IEEE
Can

1101]

2. KuAguanTe, (Ln = Ce, La) (IV, [0

Structure Description of K 2. 5Ag4, 5Ln2T e9 (Ln = Ce, La) (IV, 10 -— The two
compounds, K2_5Ag4.5Ce2Te9 and K2_5Ag4,5LazTeo, are isostructural and crystallize
in the orthorhombic space group, Immm. The two-dimensional structure, as seen
down the b-axis, is shown in Figure 4.10. Much like KCuCeTe4, the structure is
made up of two “distinct” layers that alternate in an A-B—A-B fashion and can be
regarded as an intergrowth compound. For the purpose of this discussion, the
following description will be for that of K2_5Ag4,5Ce2Teo, The first layer,
[CeTe3]°'5', also exists in KCuCeTe4 except now the layer possesses an overall
negative charge. This layer once again adopts the NdTe3 structure type with all Ce
atoms coordinated to nine Te atoms in a monocapped square anti-prismatic
arrangement, see Figure 4.11A. This layer contains a Te net that is perfectly
square with all Te-Te distances being equal at 3.1806(5)A, see Figure 4.11B. The
second layer, however, is different from that of KCuCeTe4. Instead of being made
up of a “single layer” of tetrahedrally coordinated copper atoms, as was observed
in the [CuTe] layer of KCuCeTe4, this layer is now made up of “double layers” of
tetrahedrally coordinated silver atoms that share edges in two dimensions.
Conceptually, it is as if two “single layers” of tetrahedrally coordinated silver
atoms have been sewn together. Interestingly, the Te atom that acts as the point of
connection between these “single layers” is eight coordinate, a rather high
coordination number for chalcogens. This layer alone is very similar the one that

exists in KCu4S333 and can be written as [K1.5Ag4.5Te31- The difference, however,

157

 

be»

insil

earl

morl

anis

com

is that, in K25Ag45Ce2Teo, another cation exists in-between these “double layers”.
This cation is disordered between potassium and silver (50:50). This layer can also
be viewed as being made up of individual cages where now the cation is sitting
inside the cavities of each cage. Finally, the potassium ions are stabilized between
each “distinct” layer in a square antiprismatic geometry, see Figure 4.11C. A
more descriptive way to write the formula of this compound is therefore
[(K+)(K1,5Ag4,5Te3)(CeTe3°'5')2]. The fi'actional atomic coordinates, isotropic and
anisotropic temperature factors, bond distances, and bond angles for both

compounds is given in Tables 4.17-4.20.

158

 

 

Figure 4.10 ORTEP representation of the extended structure of
K2.5Ag4,5Ce2Te9 as seen down the b-axis (90% probability ellipsoids). The ellipses

with octant shading represent Ce, the crossed ellipses represent Ag, the large open
ellipses represent K and Te.

159

 

 

Figure 4.11 ORTEP representations of (A) the coordination environment
around Ce in K25Ag45Ce2Te9 (50% probability ellipsoids), (B) the “Te net” of
K2.5Ag4,5Ce2Te9 (70% probability ellipsoids), and (C) the coordination environment
around K in K25Ag45Ce2Te9 (90% probability ellipsoids) The ellipses with octant
Shading represent Ce and the large open ellipses represent Te.

160

 

lab
Dis;

will

Kw

8103

Ce
let 1
Tel
let}
in:
145
Ag
1ng

1':

Table 4.17 Fractional Atomic Coordinates and Equivalent Isotropic

Displacement Parameters (qu) for K2,5Ag4_5Ce2Te9 (IV) and K2.5Ag4,5La2Te9 (V)
with Estimated Standard Deviations in Parentheses.

 

 

 

 

 

K2,5Ag4.5Ce2Te9
atom x y z occupancy Ueq“, A2
Ce 0.0 0.0 0.2907(1) 1.0 1.7(1)
Te(l) 0.0 0.0 0.2264(1) 1.0 1.50)
Te(2) 0.5 0.0 0.3403(1) 1.0 2.6(1)
Te(3) 0.0 0.5 0.3404(1) 1.0 2.80)
Te(4) 0.0 0.0 0.4274(1) 1.0 3.00)
Te(5) 0.5 0.5 0.5 1.0 2.9(1)
Ag(l) 0.5 0.0 0.4608(1) 1.0 4.60)
A130) 0.0 0.5 0.4607(1) 1.0 4.5(1)
1(0) 0.5 0.5 0.3985(2) 1.0 3.7(2)
K(2) 0.0 0.0 0.5 0.550) 3.4(3)
A8(3) 0.0 0.0 0.5 0.450) 3.4(3)
K2.5Ag4,5La2Te9

a 2
atom x y z occupancy Ueq , A
La 0.0 0.0 0.2912(1) 1.0 0.9(1)
Te(l) 0.0 0.0 0.2261(1) 1.0 0.9(1)
Te(2) 0.5 0.0 0.3417(1) 1.0 2.2(1)
Te(3) 0.0 0.5 0.3416(1) 1.0 2.0(1)
Te(4) 0.0 0.0 0.4273(1) 1.0 2.2(1)
Te(S) 0.5 0.5 0.5 1.0 2.4(1)
A80) 0.5 0.0 0.4607(1) 1.0 3.4(1)
A80) 0.0 0.5 0.4608(1) 1.0 3.5(1)
K0) 0.5 0.5 0.3992(2) 1.0 25(2)
K(2) 0.0 0.0 0.5 054(3) 2.2(2)
A80) 0.0 0.0 0.5 046(3) 22(2)

 

“Ueq is defined as one third of the trace of the orthogonalized Uij tensor

161

Table 4.18 Anisotropic Displacement Parameters (A) for KzisAgUCezTeo (IV)
and K2,5Ag4,5LazTe9 (V) with Standard Deviations in Parentheses.

 

 

 

 

 

£2.5Ag45C32Te9

atom Ull U22 U33 U12 U13 023
Ce 0.021(1) 0.017(1) 0.013(1) 0 0 0
Te(l) 0.019(1) 0.015(1) 0.013(1) 0 0 0
Te(2) 0.038(2) 0.023(1) 0.018(1) 0 0 0
Te(3) 0.043(2) 0.023(1) 0.017(1) 0 o 0
Te(4) 0.030(1) 0.026(1) 0.034(1) 0 0 0
Te(5) 0.036(2) 0.031(2) 0.021(2) 0 0 0
Ag(l) 0.041(2) 0.063(3) 0.034(2) 0 0 0
Ag(2) 0.065(3) 0.038(2) 0.034(2) 0 0 0
K0) 0.043(5) 0.048(6) 0.021(4) 0 0 0
K(2) 0.040(4) 0.034(4) 0.029(4) 0 0 o
Ag(3) 0.040(4) 0.034(4) 0.029(4) 0 0 0
KnggrsLazTeo

atom U11 U22 U33 U12 Ul3 U23
La 0.005(1) 0.016(1) 0.007(1) 0 0 0
Te(l) 0.004(1) 0.014(1) 0.010(1) 0 0 0
Te(2) 0.018(1) 0.038(1) 0.008(1) 0 0 0
Te(3) 0.019(1) 0.032(1) 0.008(1) 0 0 0
Te(4) 0.011(1) 0.021(1) 0.035(2) 0 0 0
Te(S) 0.028(2) 0.034(2) 0.01 1(2) 0 0 0
A80) 0.026(2) 0.044(2) 0.033(2) 0 0 0
A8(2) 0.036(2) 0.036(2) 0.033(2) 0 0 0
1(0) 0.022(4) 0.025(4) 0.027(6) 0 0 0
K0) 0.017(3) 0.027(3) 0.022(4) 0 0 0
go) 0.017(3) 0.027(3) 0.022(4) 0 0 0

 

162

Table 4.19 Selected Distances (A) and Bond Angles (deg) for
K2,5Ag4_5Ce2Teo (IV) with Standard Deviations in Parentheses.

 

 

 

Bond Distances

Ce - Tel 3.270(3), 3.2963(9) Ag2 — Te5 3.005(3)
Ce—Te2 3.376(2) Agl-Ag2 3.1806(4)
Ce-Te3 3.389(2) K1—Te2 3.722(7)
Te2 — Te3 3.1806(5) K1 — Te3 3.709(7)
Agl —Te4 2.812(3) K1 —Te4 3.505(4)
Agl — Te5 3.011(3) K2 — A32 3.015(3)
Ag2 - Te4 2.818(3) K2 — Te5 3.1806(5)
Bond Angles

Tel — Ce —Tel 74.77(5) x 4 8572(3) x 2 149.54(9) x 2
Tel — Ce -— Te2 7520(4) x 4 130.40(5) x 4 138.38(4) x 2
Tel — Ce — Te3 74.970) x 4 130.65(5) x 4 138.280) x 2
Te2 -— Ce — Te3 5609(4) x 4

Te2 — Te3 — Te2 89.653(17) x 4 90.346(17) x 4 l79.78(12) x 4

Te4 — Agl — Te4 105.78(16) x 1
Te4 - Agl — Te5 113.55(3) x 4
Te5 — Agl —— Te5 97.05(l l) x 1
Te4 — Ag2 - Te4 106.32(16) x 1
Te4—Ag2 —Te5 113.52(3) x4
Te5 - Ag2 - Te5 96.53(11) x 1

Te2 —- K1 —Te3 50.6800) x 4

Te2—Kl -Te4 86.79(7) x 4 136.40(14)x4

Te3 — Kl .— Te3 74.3907) x 1

Te3 ~K1—Te4 87.01(7) x 4 136.1604) x4

Te4—-Kl —Te4 79.5501) x 2 80.1301) x2 130.3(3)x2
Te5 -—K2 -Te5 89.65406) x2 90.34606) x2 180.00 x1

M

163

Table 4.20 Selected Distances (A) and Bond Angles (deg) for
K2_5Ag4_5LazTe9 (V) with Standard Deviations in Parentheses.

 

 

 

Bond Distances

La - Tel 3.295(3), 3.3109(9) Ag2 —- Te5 3.008(2)
La—Te2 3.414(2) Agl-Ag2 3.1938(4)
La - Te3 3.408(2) K1 — Te2 3.638(7)
Te2 - Te3 3.1938(5) Kl — Te3 3.689(7)
Agl —Te4 2.821(3) K1—Te4 3.496(4)
Agl —Te5 3.008(2) K2 —Ag2 3.004(3)
Ag2— Te4 2.821(3) K2 —Te5 3.1938(4)
Bond Angles

Tel — La —Te1 74.71(5) x 4 8588(3) x 2 149.43(10)x 2
Tel -La—Te2 75.230) x4 130.53(5) x4 138.520) x 2
Tel — La — Te3 7533(4) x 4 l30.43(5) x 4 138.56(4) x 2
Te2 — La - Te3 5583(4) x 4

Te2 - Te3 — Te2 89.855(l7) x 4 90.145(16) x 4 179.89(l3) x 4

Te4 — Agl — Te4 106.47(16) x 1
Te4 - Agl — Te5 113.33(3) x 4
Te5—Ag1—Te5 97.17(ll)x1
Te4—Ag2—Te4 106.13(l6) x1
Te4 — Ag2 — Te5 1 1334(3) x 4
Te5 — Ag2 —— Te5 97.48(11) x 1

Te2—K1-Te3 51.35(1l)x4

Te2 - Kl — Te4 8579(8) x 4 135.79(15) x 4

Te3 —K1—Te3 75.6l(18)x1

Te3 — Kl — Te4 8570(8) x 4 135.8805) x 4

Te4 — K1 — Te4 80.35(11) x 2 80.600 1) x 2 132.0(3) x 2
Te5 — K2 - Te5 89.855(16) x 2 90.145(16) x 2 180.00 x l

k

164

Transmission Electron Microscopy of K 2, 5Ag4. 5Ce2T e9 (IV) — The [CeTe30'5‘]
layer in KzsAgrsCezTeo contains a square Te net that is fillly occupied. The
formal oxidation state on the Te atoms in this net, however, is —0.75. This value
indicates that the Te net is electron deficient and therefore susceptible to
distortion. To probe the existence of such a distortion, electron diffraction studies
were performed on this material. Figure 4.12A shows a typical electron
diffiaction pattern for K25Ag45Ce2Te9 along the ab-plane which, indeed, shows
evidence for a superstructure. A densitometric intensity scan obtained fi'om the
(hkO) reciprocal plane along the (h20) row of reflections is shown in Figure 4.128.
The reflections between the (020), (120), and (220) reflections are due to a
0.333a“ superlattice, which corresponds to a 3 x asub (i.e. ~ 13.5 A) lattice
dimension. These supercell reflections also occur along the b* direction, at the
same exact position, giving a commensurate “3am x 3b,“), ” supercell. However,
as seen for KCuCeTe4, it is possible that this supercell is an artifact caused by
tWinning of the crystals underneath the electron beam (rotated 90° with respect to
one another) and the true supercell is that of a “lamb x 3am”. Long axial
photographs were taken on the x—ray diffractometer for this compound and the

results support this premise, showing supercell reflections along only one axis.

165

Figure 4.12 (A) Selected area electron diffraction pattern of
K25AgtsCe2Te9 with the electron beam perpendicular to the layers ([001]
direction) showing a twinned Ba“, x 3bsub domain (i.e.; two lamb x 3bsub supercells
that are rotated 90° with respect to one another and superimposed). (B)
Densitometric intensity scan along the b*-axis of the electron diffraction pattern of
KnggMCt-QTe9 (Fig 4.11 A) (boxed area in photograph) showing the (h 2 0)
f3111in of reflections. The three reflections from the sublattice of K2_5Ag(5Ce2Te9
are indexed. The four weak peaks are from the la x 3b superlattice.

166

 

Electron Diffraction of K25 Ag45Ce2Te9

(A) ; -*

 

 

(B) (020, (220)

 
   
 

0.333a* 020)
0.333a*

Intensity (arb. units)

 

 

 

Reciprocal Angstroms

167

Superstructure Determination of K 2_ 5Ag4_ 5Ce2T ca (110 - Because the “1am x
3b,“, ” supercell of K2_5Ag4_5Ce2Te9 was commensurate, filrther attempts were
made to collect enough crystallographic data to solve the superstructure and
elucidate the Te net distortion. The original data was collected on a Rigaku
AF CS four-circle diffi'actometer, which unfortunately was not sensitive enough to
detect such weak supercell reflections. Another crystal was therefore mounted on
the more sensitive Siemens SMART Platform CCD diffractometer using graphite
monochromatized Mo Kor radiation. The data were collected over a full sphere of
reciprocal space, up to 50° in 20. The individual frames were measured with an to
rotation of 03° and an acquisition times of 60 sec/frame. The SMART22
software was used for the data acquisition and SAINT 23 for the data extraction and
reduction. The absorption correction was performed using SADABS.24 The
structure was solved by direct methods using the SHELXTL” package of
crystallographic programs.

From the matrix flames, three equivalent monoclinic-C supercells were

found, depending on which axis was chosen as the unique axis:

 

:Supercell choice #1 Supercell choice #2 Supercell choice #3

 

a= 53.31(1)A a=51.05(l)A a= l4.129(3)A

b = 4.5318(9) A b = 135070) A b = 50.440) A

c = 13.699(3) A c = 4.4800(9) A c = 4.4492(9) A

a: 90° a=90° a=90°

B== 104.84(3)° [3 = 9491(3) B = 108.37(3)°
_______Y = 90° 7 = 90° Y = 90°

 

168

After attempting to solve the superstructure in all three cell choices, it was
concluded that only one cell choice leads to a logical solution for the
superstructure. Supercell choice #1, when applied, led to a crystallographic model
which made no chemical sense. Many of the assigned atoms were extremely close
to another and the thermal ellipsoids were either very small or very large.
Supercell choice #2, when applied, managed to give a solution that made chemical
sense. However, this superstructure solution was identical to that of the subcell.
The Te net did not exhibit any sort of distortion and the potassium and silver
cations were statistically disordered in the same exact fashion as in the subcell. In
fact, there was no obvious reason why the unit cell needed to be as large as it was.
Finally, when supercell choice #3 was applied, the structural model that was found
not only made chemical sense, but possessed a distorted Te net. Below is a
comparison of the subcell parameters to those of the correct supercell and the

vectorial relationship between the two.

 

 

 

Subcell Supercell
a = 4.4844(9) A a’ = 0.130(3) A
b=4.5116(9)A b’ = 50.44100)A
c = 50.859(10) A c’ = 4.4492(9) A
or = 90° or = 90°
B = 90° [3 = 108.37(3)°
Y = 90° 7 = 90°

Vectorial Relationship:
a’=-a+3b b’=c c’=a

169

 

par.

lat

wer
plat
mer
in I
calc
den:
C135
[hi I
is v

lip]

 

Analogous data was collected for the isostructural compound,
K2,5Ag4.5LazTe9 and the same results were found. Complete data collection
parameters and details of both structure solutions and refinements are given in
Tables 4.21.

Another complication, for both compounds, lies in the fact that the crystals
were micro-twinned. The morphology of the crystals is that of very thin square
plates. This combination allows the crystals to stack and twin very easily, in a
merohedral fashion. The crystallographic reflections were therefore overlapping
in their positions, leading to an observed electron density much higher than
calculated. This results in a poor refinement and significant residual electron
density in the Fourier map. Since it seemed impossible to find a truly “single”
crystal, attempts were made to correct for this twinning by applying a twin law to
the data. Several twin laws were tried, based on modeling how the “twinned” cell
is vectorally related to the original cell.34 The twin law that gave the best
improvement on the refinement was one which the two cells are related by a

mirror perpendicular to the ac plane:

Twin Law: a’ a -1 0 2
b’ = b 0-1 0
c’ c 0 0 1

170

By applying this twin law, the R factors (R/wR2) in the refinement dropped from
1209/3977 to 1041/3234. The Flack parameter was also refined, indicating that

42% of the crystal belonged to a fiagment defined by this twin law.

171

 

lab

crys

Dill

Table 4.21 Crystallographic Data for the “la x 3b” superstructures of

K2.5Ag4,5Ce2Te9 (IV) and K2,5Ag4,5La2Te9 (V)

 

 

KzsAgcscezTes KzsAgsLazTfi
supercell supercell
crystal habit, color plate, copper plate, copper
Diffractometer Siemens SMART Siemens SMART
Platform CCD Platform CCD
Radiation Mo-Kot (0.71073A) Mo-Ka (0.71073A)
Crystal Size, mm3 0.31 x 0.22 x 0.04 0.44 x 0.13 x 0.02
Temperature, K 293 293
Crystal System Monoclinic Monoclinic
Space Group C2/m (#12) C2/m (#12)
a. 14.130(3) ' 144310)
b. A 50.44100) 50.72800)
9, A 4.4492(9) 4.5186(9)
B, ° 108.37(3) 108.42(3)
V, A3 3009.400) 3118.7(11)
z 6 6
0mm" 22.192 21.129
indexranges -185h518 -13 $11518
-655k565 -655k564
-5 515 5 -6 515 5
sec/frame 60 30
total data 11986 9925
unique data 3544 3711
R(int) 0.0842 01585
no. parameters 136 136
final R1/wR2‘, % 1041/3234 9.84/29-29
GooF 1.076 1-039

 

 

'RI=>:(IF.I - lab/Elm w82={2(me-F3)21/21w<Foz>21}'”

172

Superstructure Description of K 2, 5Ag4, 5Ce 2 T e9 (110 - The superstructure as
seen down the c-axis is shown in Figure 4.13. Within the [K._5Ag4,5Te3] layer, the
disorder between the potassium and Silver is now partially resolved, see Figure
4.14A. While one of the crystallographic sites is now fully assigned as silver, the
other Site retains a 50/50 disorder between K and Ag. The arrangement of the
cations across the layer is now periodic in that every third cation is Ag. From
this, it is understandable why the supercell is that of a “lasub x 3bsub”. Within the
[CeTe30'5’] layer, the Te net is now distorted (as expected). A fiagment of this
layer is shown in Figure 4.143 and comparison between the Te net in the
substructure and superstructure is shown in Figure 4.15. The net is still fillly
occupied, but now the Te atoms have oligomerized into infinite zig-zag chains.
The Te-Te distances range from 2.922(3) — 3.0509(17) A within the chain and
3.2627-3.3687A between the chains. The fractional atomic coordinates, isotropic

and anisotropic temperature factors, bond distances, and bond angles for both

compounds is given in Tables 22-27.

173

 

 

 

 

 

 

 

 

K/Ag 0 0 9 Ag
0 O O (3
Te
Ce
0 o o o o o o K
O O O 0 Te
Ag
0 O O O O O O
O O O O

9 O O 9
O o O O O O
a
Figure 4.13 ORTEP representation of the “1am, x 3b“;J ” superstructure 0f

K2.5A84.5CeqTe9 as seen down the b-axis (75% probability ellipsoids). The ellipses
with octant shading represent Ce, the crossed ellipses represent Ag, the large open
ellipses represent K and Te.

174

 

(A)

 

 

 

 

Figllre 4.14 ORTEP representation (50% probability ellipsoids) of (A) The
[KisAgrsTe3] layer of the law], x 3bsub superstructure of K25Ag45Ce2Te9, and (B) a

frfigment of the [CeTe30'5] layer of the lam x Bbwb superstructure of
K2.5A84.5C02Te9 highlighting the particular coordination environment of Ce. The

ellipses With octant shading represent Ce, the crossed ellipses represent Ag, and
the large open ellipses represent K and Te.

175

 

 

F' -
K'glll’e 4.15 Vrew of the Te "nets" in (A) the substructure of
2.5A84.5CezTe9 and (B) the lamb x 3bsub superstructure of K25Ag45Ce2Te9.

176

Table 4.22
Displacement Parameters (Ueq) for the “1am x 3b,"), ” superstructure of

Fractional Atomic Coordinates and Equivalent Isotropic

K2_5Ag4,5Ce2Te9 (IV) with Estimated Standard Deviations in Parentheses.

 

 

atom x Y Z occupancy ch‘, A2
Ce(l) 0.5 0.2094(1) 0.5 1.0 1.9(1)
Ce(2) 0.3333(1) 0.2906(1) 0.8217(3) 1.0 1.8(1)
Te(l) 0.5 0.2734(1) 0.5 1.0 1.6(1)
Te(2) 0.3334(1) 0.2264(1) 0.8340(3) 1.0 1.8(1)
Te(3) 0.5 0.1597(1) 0.0 1.0 2.50)
Te(4) 0.3329(1) 0.3404(1) 0.2981(4) 1.0 2.2(1)
Te(S) 0.5 0.3406(1) 0.0 1.0 3.00)
Te(6) 0.3329(1) 0.1595(1) 0.2941(4) 1.0 2.1(1)
Te(7) 0.0 0.5724(1) 0.5 1.0 3.30)
Te(8) 0.3332(1) 0.4275(1) 0.8372(3) 1.0 3.0(1)
Te(9) 0.5 0.5 0.5 1.0 3.20)
Te(10) 0.1667(2) 0.5 0.1723(5) 1.0 2.9(1)
A80) 00 0.5392(1) 0.0 1.0 5.0(1)
A80) 0.3330(2) 0.4608(1) 0.3382(5) 1.0 4.70)
A00) 05 0.4606(1) 0.0 1.0 4.80)
A8(4) 0.1664(2) 0.5393(1) 0.6742(5) 1.0 4.7(1)
A86) 0.0 0.5 0.5 1.0 6.10)
K0) 0.5 0.3986(2) 0.50 1.0 3.50)
Km 0.1666(5) 0.6014(1) 0.180003) 1.0 47(2)
1(0) 0.3324(4) 0.5 0.8389(9) 050(2) 4.4(2)
Ag(6) 0.3324(4) 0.5 0.8389(9) 050(2) 44(2)

 

aUe‘l is defined as one third of the trace of the orthogonalized Uij tensor

177

Table 4.23
Displacement Parameters (Ueq) for the “1am x 3b,“), ” superstructure of

Fractional Atomic Coordinates and Equivalent Isotropic

K2_5Ag4,5LazTe9 (V) with Estimated Standard Deviations in Parentheses.

 

 

atom x Y Z occupancy Um”, A2
La0) 0.5 0.2093(1) 0.5 1.0 1.1(1)
La(2) 0.3331(1) 0.2914(1) 0.8232(3) 1.0 1.30)
Te(l) 0.5 0.2742(1) 0.5 1.0 1.2(1)
Te(2) 0.3332(1) 0.2262(1) 0.8340(3) 1.0 1.2(1)
Te(3) 0.5 0.1586(1) 0.0 1.0 3.1(1)
Te(4) 0.3324(1) 0.3417(1) 0.2997(5) 1.0 1.3(1)
HS) 05 0.3423(1) 0.0 1.0 3.60)
Te(6) 0.3335(1) 0.1585(1) 0.2986(5) 1.0 1.2(1)
Te(7) 0.0 0.5736(1) 0.5 1.0 2.0(1)
Te(8) 0.3328(1) 0.4273(1) 0.8377(4) 1.0 2.8(1)
Te(9) 0.5 0.5 0.5 1.0 3.2(1)
Te(10) 0.1648(2) 0.5 0.1710(6) 1.0 2.4(1)
A80) 00 0.5404(1) 0.0 1.0 2.8(1)
Ag(2) 0.3302(2) 0.4613(1) 0.3381(6) 1.0 3.80)
A80) 05 0.4611(1) 0.0 1.0 4.20)
MM) 0.1642(2) 0.5395(1) 0.6727(6) 1.0 3.3(1)
A8(5) 0.0 0.5 0.5 1.0 4.4(2)
K0) 0.5 0.4003(2) 0.5 1.0 24(2)
K(2) 0.1680(5) 0.6008(1) 0.178802) 1-0 24(2)
K(3) 0.3298(3) 0.5 0.8345(12) 053(2) 26(2)
A8(6) 0.3298(3) 0.5 0.8345(12) 047(2) 2.6(2)

 

aUeq is defined as one third of the trace of the orthogonalized Uij tensor

‘ 178

Table 4.24

Anisotropic Displacement Parameters (A) for the “lam,, x 3b,“), ”

superstructure of K25Ag45Ce2Teo (IV) with Standard Deviations in Parentheses.

 

 

atom U11 U22 U33 U12 U13 U23
Ce(l) 0.020(1) 0.016(1) 0.024(2) 0 13(1) 0
Ce(2) 0.026(1) 0.019(1) 0.014(1) 1(1) 13(1) 4(1)
Te(l) 0.022(1) 0.017(1) 0.012(1) 0 10(1) 0
Te(2) 0.022(1) 0.020(1) 0.015(1) 2(1) 10(1) 3(1)
Te(3) 0.047(2) 0.025(1) 0.003(2) 0 7(1) 0
Te(4) 0.021(1) 0.022(1) 0.026(2) 2(1) 10(1) 0
Te(5) 0.062(2) 0.023(1) 0.002(2) 0 5(1) 0
Te(6) 0.018(1) 0.023(1) 0.026(1) 4(1) 10(1) 4(1)
Te(7) 0.028(1) 0.044(2) 0.022(2) 0 0(2) 0
Te(8) 0.031(1) 0.042(1) 0.013(1) -1(1) 1(2) 5(1)
Te(9) 0.041(2) 0.040(2) 0.007(2) 0 -2(2) 0
Te(lO) 0.031(1) 0.023(1) 0.024(2) 0 -4(2) 0
A80) 0.051(2) 0.046(2) 0.039(3) 0 -3(2) 0
A8(2) 0.067(2) 0.044(2) 0.019(2) -10) -1(2) 5(1)
1‘00) 0.052(2) 0.041(2) 0.037(2) 0 -5(2) 0
A8(4) 0.036(2) 0.044(2) 0.047(2) 0 -6(2) 0
Ag(5) 0.062(4) 0.041(3) 0.061(5) 0 -8(4) 0
1(0) 0.051(5) 0.014(3) 0.028(5) 0 -6(5) 0
K0) 0.066(5) 0.015(3) 0041(5) -6(2) 40(6) 4(2)
1(0) 0.044(4) 0.055(4) 0.017(3) 44(3) 0
Ag(6) 0.044(4) 0.055(4) 0.017(3) 44(3) 0

 

179

Table 4.25 Anisotropic Displacement Parameters (A) for the “lam, x 3b,“), ”
superstructure of K2_5Ag4.5LazTeo (V) with Standard Deviations in Parentheses.

 

 

atom U11 U22 U33 U12 U13 U23
La(1) 0.012(1) 0.003(1) 0.028(2) 0 20(1) 0
La(2) 0.015(1) 0.004(1) 0.030(1) 1(1) 21(1) 2(1)
Te(l) 0.012(1) 0.005(1) 0.026(2) 0 16(1) 0
Te(2) 0.012(1) 0.005(1) 0.026(1) -10) 16(1) 2(1)
Te(3) 0.061(2) 0.003(1) 0.033(3) 0 22(2) 0
Te(4) 0.012(1) 0.006(1) 0.027(1) 1(1) 16(1) 0
Te(5) 0.059(2) 0.004(1) 0.041(2) 0 9(2) 0
Te(6) 0.012(1) 0.006(1) 0.026(1) 4(1) 17(1) 1(1)
Te(7) 0.019(1) 0.014(1) 0.034(2) 0 21(2) 0
Te(8) 0.021(1) 0.033(1) 0.040(2) -30) 25(2) 1(1)
Te(9) 0.044(2) 0.009(2) 0.049(4) 23(2) 0
Te(lO) 0.028(1) 0.007(1) 0.043(2) 22(2) 0
A80) 0.041(2) 0.016(2) 0.024(3) 8(2) 0
Ag(2) 0.040(2) 0.037(2) 0.031(2) -30) 2(2) -20)
A80) 0.037(2) 0.041(2) 0.041(4) 0 4(3) 0
A80) 0.029(1) 0.026(1) 0.042(2) -50) 11(2) -90)
A8(5) 0.055(3) 0.031(3) 0.059(6) 0 39(4) 0
K0) 0.018(4) 0.025(5) 0.013(5) 0 -20(4) 0
K0) 0.031(3) 0.012(3) 0.010(2) 2(2) -20(6) 6(2)
1(0) 0.025(3) 0.016(3) 0.049(5) 0 27(3) 0
Ag(6) 0.025(3) 0.016(3) 0.049(5) 0 27(3) 0

 

180

Table 4.26 Selected Distances (A) and Bond Angles (deg) for the “1am, x

3b,..b ” superstructure of K25Ag45Ce2Teo (IV) with Standard Deviations in
Parentheses.

 

Bond Distances

Cel — Tel 3.226(3)

Cel -- Te2 3.2370( 14), 3.2644(18) Ag4 — Te7 2.786(3)
3.269305), 3.306308) Ag4 — Te8 2.794(3)

Cel — Te3 3.353(2) Ag4 — TelO 2.974(3), 2.989(3)

Cel — Te6 3.377(2) Agl — Ag4 3.126(3). 3.175(2)

Ce2 — Tel 3.2370(14), 3.3062(18) Ag2 — Ag3 3.140(2), 3.173(3)

Ce2 — Te2 3.234(2), 3.240(2) Ag2 — Ag4 3.145(3), 3.162(4)
3.3014(18) K1 - Te4 3.696(6)

Ce2 — Te4 3.286(2), 3.422(2) K1 — Te5 3.675(6)

Ce2 — Te5 3.368(3) K1 - Te8 3.464(4), 3.487(4)

Ce2 — Te6 3.369(2) K2 —. Te3 3.694(7)

Te3 — Te6 3.0391(18) K2 — Te4 3.696(6)

Te4 — Te5 3.05090 7) K2 — Te6 3.612(6), 3.748(7)

Te4 — Te6 2.922(3) K 2 — Te7 3.437(7), 3.512(6)

Agl — Te7 2.784(3) K2 — Te8 3.446(6), 3.500(8)

Agl — Te10 2.986(3) Ag5 — Agl 2.977(3)

Ag2 — Te8 2.784(3). 2.793(3) Ag5 — T e10 3.137(2)

Ag2 — Te9 2.987(3) K3/Ag6 —Ag2 2.972(4)

Ag2 — TelO 2.980(3) K3/Ag6 —Te9 3.141(4)

Ag3 - Te8 2.793(3) K3/Ag6 —Te10 3.142(6)

Ag3 - Te9 2.982(3)

 

 

 

181

Table 4.26 continued Selected Distances (A) and Bond Angles (deg) for the

“1am x 3b,”, ” superstructure of K25Ag45Ce2Te9 (IV) with Standard Deviations in
Parentheses.

 

Bond Angles

Tel — Cel —Te2 74.84(4) Te9 - Ag2 — Te10 96.99(10)

Tel - Cel - Te3 138.430) Te8 — Ag3 — Te8 106.4506)

Tel -- Cel —- Te6 138.280) Te8 — Ag3 - Te9 113.810)

Te2 — Cel — Te2 85.850), 149.69(8) Te9 — Ag3 - Te9 96.4702)

Te2 — Cel - Te3 75.200), 13037(5) Te7 — Ag4 — Te8 106.3803)

T62 - Cel — Te6 7294(5), 127.88(5) Te7 — Ag4 - Te10 114.2200)

Te3 — Cel - Te3 8313(8) Te8 — Ag4 — Te10 113.78(9)

Te3 — Cel — Te6 5369(5) Te10 — Ag4 - Te10 96.53(9)

Te6 - Cel - Te6 8344(7) Te4 — K1 — Te4 74.8405)

Tel — Ce2 - Te2 7509(5), 149.3(7) Te4 - K1 — Te5 52.63(10)

Tel - Ce2 — Te4 7594(5), 131.42(6) Te5 — K1 — Te5 74.5005)

Tel — Ce2 - Te5 7475(5) Te5 - K1 — Te8 8677(7), 136.2402)
Tel - Ce2 — Te6 126.69(6),134.59(6) Te8 — K1 — Te8 79.5900), 130.2(2)
Te2 — Ce2 - Te2 7426(5), 8580(5) Te3 — K2 — Te4 74.3802)

Te2 — Ce2 — Te4 7470(5), 138.82(5) Te3 — K2 — Te6 4914(8)

Te2 — Ce2 — Te5 129.29(5) , 138.34(6) Te3 — K2 — Te7 87.4607)

Te2 — Ce2 — Te6 77.66(6), 138.27(6) Te3 — K2 — Te8 137.6109)

Te4 — Ce2 — Te4 8307(6) Te4 — K2 — Te6 4713(9)

Te4 — Ce2 — Te5 54.560) Te4 - K2 — Te7 132.59(7)

Te4 — Ce2 — Te6 5208(6), 5978(5) Te4 — K2 — Te8 8893(15)

Te5 - Ce2 -- Te6 8306(6) Te6 — K2 - Te6 74.37(12)

Te7 - Agl — Te7 106.06(17) Te6 - K2 — Te7 87.6203), 135.6(2)
Te7 - Agl _ Te10 11 390(5) Te6 — K2 — Te8 87.5505), 135.4(2)
Te10-Ag1—Te10 9699(14) Te7-K2—Te7 79.6104)

Te8 — A32 — Te8 105.8403) Te7 - K2 — Te8 809603), 1303(2)
Te8 — Ag2 _ Te9 113320;) Te8 — K2 — Te8 79.6605)
Te8—Ag2 —Te10 113.6100) Te6—Te4 —Te5 9579(7)

 

 

 

182

Table 4.27 Selected Distances (A) and Bond Angles (deg) for the “1am, x
3b,.n,” superstructure of K2_5Ag4,5La2Te9 (IV) with Standard Deviations in

 

Parentheses.

Bond Distances

Lal — Tel 3.293(3)

Lal - Te2 3.293(3), 3.312(2) Ag4 — Te7 2.824(3)
3.204(15) Ag4 - Te8 2.845(4)

Lal - Te3 3.422(3) Ag4 - Te10 3.013(3), 3.029(3)

Lal - Te6 3.436(2) Agl — Ag4 3.151(2). 3.203(3)

L82 — Tel 3.2916(14), 3.349(2) Ag2 — Ag3 3.205(3), 3.258(3)

La2 — Te2 3.291(2), 3.306(2) Ag2 — Ag4 3.192(4), 3.199(3)
3.3487(18) K1 - Te4 3.696(6), 3.747(8)

La2 — Te4 3.343(2), 3.479(2) K1 - Te5 3.709(8)

La2 — Te5 3.438(3) Kl — Te8 3.471(4), 3.503(4)

L32 - Te6 3.414(2) K2 — Te3 3.718(7)

Te3 — Te6 3.0910(17) K2 - Te4 3.680(6)

Te4 ._ Te5 3.1100(17) K2 — Te6 3.630(6), 3.763(6)

Te4 _ Te6 2.934(3) K 2 — Te7 3.467(6), 3.528(6)

Agl - Te7 2.818(3) K2 — Te8 3.469(6), 3.507(7)

Agl — Te10 3.038(3) Ag5 — Agl 3.050(2)

Ag2 — T88 2.831(4). 2.852(3) Ag5 - TelO 3.165(2)

Ag2 — Te9 3.033(3) K3/Ag6 —Ag2 2.983(4), 3.005(5)

Ag2 - Te10 2.987(4) K3/Ag6 -Te9 3.219(5), 3.252(4)

Ag3 — Te8 2.848(4) K3/Ag6 —Te10 3.179(6), 3.191(4)

Ag3 — Te9 3.001(3)

 

 

183

Table 4.27 continued Selected Distances (A) and Bond Angles (deg) for the
“1am x 3b“), ” superstructure of K2,5Ag4_5La2Te9 (IV) with Standard Deviations in

 

Parentheses.

Bond Angles

Tel — Lal —-Te2 7495(4) Te9 — Ag2 — Te10 98.5000)

Tel - Lal - Te3 138.68(4) Te8 — Ag3 -— Te8 106.06(17)

Tel - Lal -— Te6 138.640) Te8 — Ag3 - Te9 113.72(5)

Te2 - Lal — Te2 85.890), 14990(9) Te9 — Ag3 - Te9 97.6903)

Te2 — Lal — Te3 75.280), 13010(5) Te7 — Ag4 — Te8 106.0202)

Te2 —Lal -Te6 73.28(5),132.65(6) Te7 —Ag4 -Te10 114.79(9)

Te3 — Lal — Te3 8264(7) Te8 — Ag4 — Te10 113.33(11)

Te3 — Lal — Te6 5358(4) Te10 — Ag4 — Te10 9680(9)

Te6 — La 1 - Te6 8272(7) Te4 — K1 — Te4 75.1708)

Tel — La 2 — Te2 7321(5), 0913(7) Te4 — K1 — Te5 5278(12)

Tel - La 2 - Te4 7597(6), l29.72(6) Te5 — K1 - Te5 75 .05(1 8)

Tel — La 2 - Te5 74.88(5) Te5 — K1 — Te8 8549(8), 134.870 5)
Tel — La2 - Te6 127.10(6), 134.1 1(6) Te8 — K1 — Te8 80.75(11), 133.4(3)
Te2 — La2 - Te2 7429(5), 8578(5) Te3 — K2 — Te4 7541(12)

Te2 - La2 — Te4 7603(5), 138.99(6) Te3 — K2 — Te6 4974(9)

Te2 — La2 - Te5 129.65(5), l38.6l(6) Te3 — K2 — Te7 84.38(13)

Tc2 — La2 — Te6 7510(5), 138.l3(6) Te3 — K2 — Te8 l36.85(l7)

Te4 — La 2 - Te4 8294(5) Te4 — K2 - Te6 4818(9)

Te4 —- La 2 — Te5 5725(5) Te4 — K2 — Te7 l31.94(17)

Te4 — La2 - Te6 5240(6), 5945(5) Te4 — K2 — Te8 88.13(14)

Te5 -- L32 - Te6 8501(6) Te6 — K2 — Te6 7534(12)

Te7--Agl -—Te7 106.5804) Te6 — K2 -—Te7 84.49(14),135.3909)
Te7 — Agl — Te10 114.20(5) Te6 — K2 — Te8 86.8304), 135.7008)
Te10-Agl —-TelO 95.1502) Te7—K2-Te7 8047(13)

Te8 — Ag2 — Te8 105.3304) Te7 — K2 - Te8 81.5003), 132.5(2)
Te8 —- Ag2 - Te9 1 1324(10) Te8 — K2 — Te8 80.73(l4)

Te8 ~—Ag2 —Te10 113.9001) Te6 —-Te4 —Te5 9627(8)

k

 

184

-_‘__‘....' - 4.“... h 4.“.

Magnetic Susceptibility and Infiared Spectroscopy of K 2, 5Ag4, 5Ce 2T e9 (IV)
The magnetic susceptibility of K25Ag45Ce2Te9 was measured over the range 5-
300K at 6000G. A plot of l/xM vs T Shows that the material follows Curie-Weiss
Law with only slight deviation from linearity beginning below 50K, see Figure
4.16A Such deviation has been reported for several Ce3+ compounds and has been
attributed to crystal field splitting of the cation’s 2F5/2 ground state.32 The data was

fillther corrected for Pauli paramagnetism by applying a X'np value of 0.0005

emu/mo]. A straight line curve fit to the data above 60K gives a he“ of 2.63 1.13,

which is in accordance with the usual range for Ce3+ compounds (2.3-2.5 03). The

Weiss constants, 0, was calculated to be —31K, suggesting a certain amount of
antiferromagnetic ordering. The diffuse reflectance optical Spectra was taken in
the Mid—IR region for K2,5Ag4,5Ce2Te9, see Figure 4.16B. Since the Te net in this
compound is clearly distorted, the expected behavior is that of a semiconductor.
Indeed, that is what is observed; an abrupt optical gap is observed at 0.26 eV,
which can loosely be characterized as its bandgap. An analogous spectrum for
K2.5Ag4,5LazTeo (not shown here) was obtained which conversely indicates that
this material is either a semimetal or a metal. A small optical gap is observed
around 0.05-0.07eV. However, the spectroscopic data are unreliable in this

region.

185

l/xM

350
300
250
200
150

50

a/S (arbitrary units)
.4..."..............nrr..1.

5.9—5.52- 5- -..-

 

 

 

 

- (A) .
O
t - '
O
l- O
O
- O
O
- o. O . .
L ...
h- ..
’ l l l I I
0 50 100 150 200 250 300
Temperature (K)

 

 

IIIII'TI'InI'IIII'IIII'II'IIIIII'IIT

(B)

0

LIJILJI

  

IlllJllllll

Ill.-llllJllLlJLlllllllllLllJ

 

 

O

0.1

0.2 0.3 0.4 0.5
Energy (eV)

0.6 0.7

FD
00

Figure 4.16. (A) Inverse molar magnetic susceptibility (l/xM) plotted against
temperature (2-300K) for K2_5Ag4.5Ce2Te9. (B) Diffuse reflectance optical spectra
of K2.5Ag4.5Ce2Te9 (in the Mid-1R region).

186

E10

mat

C011

bod

and

till

cry:

mill

Charge Transport Measurements of K 2, 5Ag4. 5Ln 2T e9 (Ln = Ce, La) (IV, 10
Electrical conductivity and thermopower measurements were made on both
materials. The data agree with the semiconducting behavior indicated by the Mid-
IR diffuse reflectance measurements; the electrical conductivity for both
compounds decreases with decreasing temperature. Measurements were made on
both a room-temperature pressed pellet and a single crystal of the cerium analog
and the results are shown in Figure 4.17A. The room temperature conductivity
values are not far apart, ranging fiom ~13 S/cm for the pellet and ~29 S/cm for the
crystal. In a normal polycrystalline pressed pellet, the conductivity values can be
suppressed to as little as 1/100‘h of the actual values due to the existence of grain
boundaries. However, the pellet used here was a compact of small crystals and
not of a polycrystalline powder. Therefore, the number of grain boundaries are
minimized. The thermopower data for the cerium analog is shown in Figure 4.178
and gives a room temperature Seebeck coefficient of ~130 uV/K. Therefore, the
material is best described as a p-type semiconductor.

Analogous data was collected for a room temperature pressed pellet of the
La analog and the results are Shown in Figure 4.18. The room temperature
conductivity values for the two pellets were 0.3 and 14 S/cm. This is in range with
what was observed for the Ce analog. The thermopower data for a pressed pellet
of the La analog is shown in Figure 4.18B. While the magnitudes are very similar
to that of the cerium analog (~170 mV/K at 300K), the s10pes are somewhat

different. For the lanthanum analog, there is a reproducible convex dip in the data

187

around 170K which does not exist for the cerium analog. The origin of this dip iS

yet unknown.

188

Electrical Conductivity
(A)

 

 

 

 

 

 

0 50 100 150 200 250 300

 

 

 

 

 

Temperature(K)
Thermo wer
140 p0
. B
120 -( )
E
100 _-
A )-
¥ 801
l: :
v 60;-
m I
4o:-
20 E- 0..
:f
0 lelllllljjllLllklllelJlJL
0 50 100 150 200 250 300

Temperature (K)

Figure 4.17 (A) F our-probe electrical conductivity data of both a room
temperature pressed pellet and crystal of K25Ag45Ce2Teo as a filnction of
temWrature. (B) Thermopower data of a crystal of K2,5Ag4_5Ce2Teo as a function
of temperature.

189

 

100.00

10.00

8

Log 0 (S/cm)

0.01

0.00

200

Electrical Conductivity

 

 

 

 

 

so 00......”
O
«I . .
O.

..a /
ILL_LIJ_lLlLJ-_lellelUlJl-ALLLA-Jl
0 50 100 150 200 250 300 350
Temperature (K)

Thermopower

(B) I

l

1

4

..,.::: 1

0°:::°..‘ .1

:1" 1

1

+

d

J

o ‘ :

llLlllllllllllllll.l_l_ I_LJ_L
0 50 100 150 200 250 300

Temperature (K)

Figure 4.18 (A) F our-probe electrical conductivity data of a room temperature

pressed pellet of K2,5Ag4.5LazTeo as a function of temperature. (B) Thermopower

data of a room temperature pressed pellet of K2,5Ag4,5La2Teo as a function of

temperature.

190

is 51
c001
[till
11)):
equz
1).
lion
and
the I
call
1110
Orin
on l
(011

flag

3- CaucaEuTez (VI)

Structure Description of CupnguTez (VI) -— The structure of CumsEuTez
is shown in Figure 4.19. It adopts the CaMnBiz structure-type and features 8-
coordinate europium atoms in a square-antiprismatic coordination environment of
tellurium. The europium atoms are sandwiched between a [CuTe]' anti-PbO type
layer and a flat square net of Te atoms. The Te -Te distances in the net are all
equal at 3.168(1)A, a value substantially longer than the normal Te-Te bond of 2.8
A yet much shorter than the van der Waals contact of 3.8—4.0 A. The fractional
atomic coordinates, isotropic and anisotropic temperature factors, bond distances,
and bond angles for both compounds is given in Tables 28—30. The bonds around
the europium atoms have been omitted in Figure 4.19 to highlight the stacking of
each individual layer. The copper site is only partially occupied and refines to a
value of 0.66. Since partial occupancy on copper is unusual and the crystals were
originally isolated from a szTex flux, careful elemental analysis was performed
on the single crystal after data collection was complete. This analysis not only
confirmed the absence of rubidium, but verified the copper composition to be
exactly 0.66. Interestingly, this structure type has been encountered in the family
of antirnonides MxLasz aw: Zn, Co, Mn, and Cu; x=0520.87).” In all of these
phases, the transition metal site is partially occupied (although the reason for this
has yet to be addressed). CUO_66EUT@2 appears to be the first telluride member of

this family.

191

 

 

Figure 4.19 ORTEP representation of the structure of CuomEuTez as seen down
the b—axis (70% ellipsoids). The ellipses with octant shading represent Eu atoms.

The crossed ellipses represent Cu atoms and the open ellipses represent Te atoms.

192

 

Tat
Dis]
Del

8101)

Eu
Te(T

Cu

Ill

5111

Table 4.28
Displacement Parameters (UN) for Cu0_66EuTe2 (IV) with Estimated Standard

Deviations in Parentheses.

Fractional Atomic Coordinates and Equivalent Isotropic

 

 

atom x y z occupancy Ueq’, A2
Eu 0.25 0.25 0.7402(5) 1.0 2.5(2)
Te( 1) 0.75 0.25 0 1.0 2.6(2)
Te(2) 0.25 0.25 0.6417(6) 1.0 2.2(2)
Cu 0.25 0.25 0.5 0.66 3.0(4)

 

'Uq is defined as one-third of the trace of the orthogonalized Uij tensor.

Table 4.29 Anisotropic Displacement Parameters (A) for CunssEuTez (IV) with
Standard Deviations in Parentheses.

 

 

atom Ull U22 U33 U12 U13 U23
Eu 0.021(2) 0.021 (2) 0.033(3) 0 0 0
Te(l) 0.030(2) 0.030(2) 0.018(4) 0 0 0
Te(2) 0.019(2) 0.019(2) 0.027(5) 0 0 0
Cu 0.036(6) 0.036(6) 0.019(9) 0 0 0

 

193

 

Ta

wit

Eu

Tel

Cu

Bo

Te}

Tel

lei

lei

lei

Table 4.30 Selected Distances (A) and Bond Angles (deg) for CuonguTez (IV)
with Standard Deviations in Parentheses.

 

 

 

Bond Distances

Eu—Tel 3.482(4) x 8

Eu - Te2 3.326(3) x 7

Tel -Te1 3.1685(11) x4

Cu — Te2 2.671(4) x 6

Bond Angles

Tel —Eu-—Tel 54.13(7)x4 80.10(11)x2
Tel -Eu—Te2 78.4200) x 7 131.78(12)x 8
Te2 -Eu—Te2 8469(8) x4 144.6(3) x 2
Tel - Tel - Tel 90.00(O) x 4 180.00(0) x 2
Te2—Cu—Te2 107.23(11)x4 114.1(2)x2

 

194

 

KC

bee:

SIIU

the

3101

tilt

the

151

pla

ch

150

OKV
A

me

till

C01

Sq:

Its

4. A,M(3_,,EuTe4 (VII, VIII)

Structure Description of A,M(3.x)EuTe4 (VII, VIII) — The structure of
KCquuTea is actually polar, see Figure 4.20. The bonds to europium have now
been included to highlight its square antiprismatic coordination environment. The
structure of KCU2EUTC4 can be derived fiom that of CuOMEuTez by first restoring
the Cu site to full occupancy and then replacing every other layer of europium
atoms with potassium. This replacement of atoms is reasonable if we compare the
effective ionic radii of Eu” (1.17A) to that of K+ (1.38A). Thus, we can expect
the europium to be truly divalent since a trivalent europium (ionic radius 0.947A)
is too small for such a site. As a result, this replacement has caused the n-glide
plane to be lost in moving from Cuo_66EuTe2 to KCquuTea and the symmetry to
change from centrosymmetric to non-centrosymmetric. Nao_2Ag2,3EuTe4 is
isostructural to KCquuTea, but now there exists some disorder on the Na site with
Ag, which can be explained again by comparing the effective ionic radii of the two
metals. Na+ and Ag+ have an ionic radii of 1.02A and 1.15A, respectively, similar
enough to allow the two metals to occupy the same site, which is also square
antiprismatic, see Figure 4213. The fractional atomic coordinates, isotropic and
anisotropic temperature factors, bond distances, and bond angles for both
compounds are given in Tables 31-34.

The Te nets in both KCquuTe4 and NaopAgnguT‘ea appear perfectly
square with all shortest T e-Te distances at 3.1371(4)A and 3.1497(4)A,

respectively, see Figure 4.21C. This can be an artifact, however, since we know

195

that

obs:

elect

 

that superstructure modulations (i.e., charge density waves) are frequently
observed in compounds with perfectly square net of atoms and are usually

electronically driven.'3’ 36

196

 

 

 

 

Figure 4.20 ORTEP representation of the structure of KCquuTea (70%
ellipsoids) viewed down the b-axis. The ellipses with octant shading represent Eu
atoms, the crossed ellipses represent Cu atoms and the open ellipses represent Te

and K atoms. The Te3 atoms make the square Te net.

197

 

1e2<

 

 

T63 Te3

 

Figure 4.21 ORTEP representation (80% probability ellipsoid) of (A) the
coordination environment around Eu in KCquuTea, (B) the coordination
environment around K in KCquuTer, and (C) a view perpendicular to the Te net

in KCquuTe4. The ellipses with octant shading represent Eu, and the large Open
elliPSes represent K and Te.

198

 

Table 4.31 Fractional Atomic Coordinates and Equivalent Isotropic
Displacement Parameters (Ueq) for KCquuTer (VII) and NaopAgnguTea (VIH)
with Estimated Standard Deviations in Parentheses.

 

 

 

 

 

KCquuTe4

atom x y z occupancy Ueq‘, A2
Eu 0.0 0.0 0.0000(3) 1.0 2.2(1)
Te(l) 0.0 0.0 0.3661(4) 1.0 2.5(1)
Te(2) 0.5 0.5 0.0940(1) 1.0 2.3(1)
Te(3) 0.5 0.0 0.7655(2) 1.0 2.4(1)
Cu 0.5 0.0 0.2416(6) 1.0 4.1(1)
K 0.5 0.5 0.5029(18) 1.0 23(2)
N30.2Ag2,3EuTe4

atom x y z occupancy Ueq', A2
Eu 0.0 0.0 0.0012(6) 1.0 37(2)
T90) 00 0.0 0.4339(4) 1.0 1.2(1)
Te(2) 05 0.5 0.1014(5) 1.0 1.0(1)
Te(3) 0.5 0.0 0.7729(7) 1.0 1.6(1)
A80) 05 0.0 0.2768(8) 1.0 1.9(1)
Ag(2) 0.5 0.5 0.5335(5) 0.79 06(2)
Na 0.5 0.5 0.5335(5) 0.21 0.6(2)

 

“Us. is defined as one-third of the trace of the orthogonalized Uij tensor.

199

Table 4.32 Anisotropic Displacement Parameters (A) for KCquuTe4 (VII) and

NaopAgnguTea (VIII) with Standard Deviations in Parentheses.

 

 

 

 

 

KCquuTea

atom Ull U22 U33 012 U13 U23
Eu 0.016(1) 0.016(1) 0.034(2) 0 0 0
Te(l) 0.020(1) 0.020(1) 0.035(2) 0 0 0
Te(2) 0.019(1) 0.018(1) 0.032(2) 0 0 0
Te(3) 0.021(1) 0.021(1) 0.031(2) 0 0 0
Cu 0.032(3) 0.065(4) 0.027(3) 0 0 0

K 0.016(3) 0.016(3) 0.036(5) 0 o 0
NaozflsEUTW

atom U11 U22 U33 U12 U13 U23
Eu 0.024(2) 0.024(2) 0.064(5) 0 0 0
Te(l) 0.012(2) 0.012(2) 0.013(3) 0 0 0
Te(2) 0.001(1) 0.001(1) 0.028(3) 0 0 0
Te(3) 0.014(2) 0.016(2) 0.019(2) 0 0 0
A80) 0.020(2) 0.020(2) 0.016(2) 0 0 0
A8(2) 0.009(2) 0.009(2) 0.000(3) 0 0 0
Na 0.009(2) 0.009(2) 0.000(3) 0 0 0

 

200

 

Tal

will

6
Eu-

Eu -
Te3

 

Table 4.33 Selected Distances (A) and Bond Angles (deg) for KCquuTea (VII)
with Standard Deviations in Parentheses.

 

 

 

Bond Distances

Eu — Te2 3.314(2)

Eu - Te3 3.467(4)

Te3 — Te3 3.1371(4)

Cu—Tel 2.631(5)

Cu -— Te2 2.781(5)

K - Tel 3.501(9)

K - Te3 372(2)

Bond Angles

Te2 — Eu —Te2 8403(7) x 4 1423807) x 2
Te2 - Eu — Te3 79.61(7) x 8 l32.53(8) x 8
Te3 — Eu - Te3 5579(6) x 4 79.55(10) x 2
Te3 — Te3- Te3 9000(0) x 4 180.00(0) x 2
Tel —Cu—Te1 114.9(3) x1

Tel — Cu — Te2 108.93(6) x 4

Te2 - Cu - Te2 105.8(3) x 1

Tel — K — Tel 786(3) x 4 127.3(6) x 2
Tel - K - Te3 88.76(15) x 8 137.3(3) x 8
Te3 —K-Te3 499(2) x4 73.2(4)x2

 

201

Table 4.34 Selected Distances (A) and Bond Angles (deg) for NaopAnguTea
(VIII) with Standard Deviations in Parentheses.

 

 

 

Bond Distances

Eu — Te2 3.341(3)

Eu—Te3 3.375(9)

Te3 - Te3 3.1497(4)

Agl —Te1 2.829(8)

Agl — Te2 2.959(8)

Agl -Ag1 3.1497(4)

Na — Tel 3.338(2)

Na— Te3 3.468(9)

Bond Angles

Te2 — Eu —Te2 83.63(8) x 4 141.1(3) x 2
Te2 - Eu — Te3 79.07( 12) x 8 133.65(13) x 8
Te3 — Eu — Te3 556(2) x 4 82.6(3) x 2
Te3 — Te3— Te3 9000(0) x 4 180.0(6) x 2

Tel -Ag1 ~Te1 103.9(4) x1
Tel — Agl — Te2 113.95(4) x 4
Te2 — Agl — Te2 97.7(4) x 1

Tel - Na —- Tel 8370(6) x 4 141.3(2) x 2
Tel —Na—Te3 7996(10) x 8 l33.03(10)x 8
Te3 — Na — Te3 54.0(2) x 4 799(3) x 2

202

 

pll
we
51;
sh
pla
bt
clo
H0
mll
bt
pm
101

ll.

all
p10

in

Si);

the

"‘IL'—_—a_,-.2....-.e."‘_-_ -— -- ~.

Transmission Electron Microscopy of A,M(3.,)Eu Te4 (VII, VIII) —- In order to
probe for a Te net distortion, we examined both KCquuTea and Nao_2Ag2_gEuTe4
were both examined by electron diffi‘action and found evidence for a
superstructure arising fi'om a distortion within the square Te net. Figure 4.22A
shows a typical electron diffraction pattern for KCquuTea depicting the (hkO)
plane. The weak spots that appear in this micrograph occur along both the a* and
b* direction and correspond to a 0.286a* x 0.286b* superlattice. This value is
close to (2/7) and therefore the supercell can be modeled as 7am x 7bsub.
However, many of the crystals examined under the electron beam were highly
microtwinned and although the modulation seems to appear along both the a* and
b* axes, it is unlikely that the superlattice is that of a 7am, x 7bsub. This pattern

probably arises from the superimposition of two lasub x 7bsub patterns that are

rotated 90° with respect to one another, as has been found for Ko.33Bao,67AgTe2,37’38

The electron diffi’action pattern of NaopAnguTea, shown in Figure 4.22B, was
taken from a very thin region of a single crystal and contains superlattice spots
along only one direction. Due to the tetragonal symmetry of the subcell, the
propensity of these crystals to twin is seemingly high and a micrograph of this sort
is difficult to obtain, since most showed superlattice spots along both the a“ and
b* directions. The spots in this micrograph correspond again to a la x 7b
superlattice and it can therefore be concluded that both of these compounds exhibit
the same superlattice. Figure 4.22C is a densitometric intensity scan along the b*-

axis of the electron diffraction pattern of NaopAnguTea (Fig 4.223). The three

203

reflections from the tetragonal sublattice are indexed. The four weak peaks are

fi'om the 7-fold supercell along this axis.

204

!--I- "‘r-__'-—‘

Figure 4.22 (A) Selected area electron diffraction pattern of KCquuTe4 with the
electron beam perpendicular to the layers ([001] direction) showing a twinned
7asub X 7bsub domain (i.e.; two lam, x 7bsub supercells that are rotated 90° with
IeSpect to one another and superimposed). (B) Selected area electron diffraction
Pattern of Na0.2Ag2,8EuTe4 with the electron beam perpendicular to the layers
([001] direction) showing the la x 7b superlattice of single crystal region. (C)
Densitometric intensity scan along the b*-axis of the electron diffraction pattern of
Nao.2Ag2_gEuTe4 (Fig 4.2113) (boxed area in photograph) showing the (-3 k 0)
f11111in of reflections. The three reflections from the sublattice of Na0_2Ag2.3EuTe4
are indexed. The four weak peaks are from the la x 7b superlattice.

205

 

Electron Diffraction of KCu 2EuTe4
and Nam Ag 2. 8 EuTe4

superlattice
sublattice

 

 

 

 

 

 

Reciprocal Angstrom:

206

 

V2

01'

10!

sh,
v (

111'

Magnetic Susceptibility Measurements of AxM3.x)EllTe4 (VII, VIII) —
Variable temperature magnetic susceptibility data for KCquuTe4 was measured
over the range of 5-300K at 6000G. A plot of l/xM vs T (see Figure 4.23A) shows
that this material exhibits perfect Curie-Weiss behavior. A um value of 7.58 BM
and a Weiss constant of ——75K was estimated by fitting a straight line to the data
above 140K. Analogous data collected for NaopAgnguTer at 3000G gave a negof
8.59 BM and a Weiss constant of -4K, see Figure 4.23B. These values are
consistent with an 1‘7 configuration or Eu” (7.9 - 8.0 BM) and are very different
from that expected for Eu3+ (3.3-3.5 BM).32

The issue of charge balancing in all three cases is anything but trivial. The
nonstoichiometry in Cu0_66EuTe2 must be taken into consideration, in addition to
the superstructures of KCquuTe4 and NaozAglgEUTCrt, when assigning formal
charges. Since the actual superstructures have not yet been determined, only the
average charge per tellurium atom in the net can be calculated, assuming Cu+, Ag”,
and Eu”. For Cuo_66EuTe2, it is best to keep the structure in mind when balancing
the charges. Since the structure is described as the packing of three layers, the
formula can best be described as [(Cu+0_66Te2')(Eu2+)(Te'°‘66)]. The formal charges
on KCquuTe4 and NaopAnguTea can be assigned using the same approach;
[(K*)(Curc)2(Eu2*)(Te*’-5)21 and [(Nah.24gb.r)(AgTe)2(Eu2*)(Ie‘°-5)). Based on
these formulations, the average charge per Te atom in the square net changes from

-0.66 in Cup-“EuTe; to -O.5 in KCquuTer.

207

l/‘v

pL_h_._.- ,~~ -_. _

 

50

40)-

l/x
f

20

10 " .0.

 

L

 

 

0 L L m L

0 50 100 150 200
Temperature (K)

250

300

 

35
30 )-
25 p
20 - .

we .

10- ’

l/xM

 

0 ’ l l l

0 50 100 150 200

Temperature (K)

250

 

J

300

Figure 4.23 Inverse molar magnetic susceptibility (1/ 1M) plotted against
temperature (2-300K) for (A) KCquuTe4 and (B) NaopAgnEuTea.

208

 

0P

Infiared Spectroscopy of AxM3.x)EuTe4 (VII, VIII) — The optical properties of
KCquuTe4 (V), and NaopAguEuTea (VII) were determined by measuring the
diffuse—reflectance spectra of each in the Mid-IR region (6000-400 cm"), see
Figure 4.24. The spectra of KCquuTe4 shows no transitions, indicating that this
material is metallic. The spectrum of NaopAguEuTen however, reveals an abrupt

optical transition at 0.24 eV, suggesting the material to be semiconducting.

 

lr'fi'fiij'riIIIUYYI—IUUYIT—r'IrIIII'I'U—I

a/S (arbitrary units)

mII‘IrFUT'IT—UI'T—‘m'rt'm

 

 

 

llllIJIJULIIIILIUIlenlIII

Ill-Jill.Ill-JlllllllljllllJJllllllllll

0 0. 1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Energy (eV)

 

Figure 4.24 Diffuse reflectance optical spectra (in the Mid-IR region) of
N30.2A82.8E11TC4-

209

“13*" ’ ' "" '\ ~5\-'A 1

 

Charge Transport Properties of A,M(3-,)EuTe4 (VII, p711) _ Electrical
conductivity and thermoelectric power data were measured as a function of
temperature for pressed pellets of KCquuTea and NaopAgnguTen see Figure
4.25. For KCquuTe4, the data suggest p—type metallic or semimetallic behavior
with a room temperature conductivity of 165 S/cm and a Seebeck coefficient of

+23 uV/K This agrees with the fact that no optical bandgap was detected for this

material in the region 0.1 to 1.0 eV. The data for Na0_2Ag2_3EuTe4 suggest p-type
semiconducting behavior with a room temperature conductivity value of 12 S/cm

and a Seebeck coefficient of +70 uV/K.

210

Electrical Conductivity

 

__-

 

100.00

 

 

 

 

 

 

 

   
 

 

 

KCu EuTe
2 4
A
8
a 10.00 4
b NaMAnguTe4
on
o
'4 1.00
0_107H34 .LaL...nln.4.J...4Lr.4.r
0 50 100 150 200 250 300
Temperature (K)
70 Thermgpower
03) .
60 - 1
E 1
50 'r '1
g i NamAnguTe4 1
> 40 E' 1
5': g :
1
V1 30 f 1
20 E g i
E E
10 E’. KCquuTe4-3
:- 00W ‘
04.1.144..1 . 1 .1...Q.L.L‘
0 50 100 150 200 250 300
Temperature (K)

Figure 4.25 (A) Four probe, electrical conductivity data of room temperature

pressed pellets of KCquuTe4 and NaopAgstuTea as a function of temperature.

(13) Thermopower data of room temperature pressed pellets of KCquuTe4 and

NaozAgstuTea as a function of temperature.

211

5- K065Ang “1.35394 (IX)

Structure Description KagAngumTerr (IX) —- The structure of
KogsAngumsTea viewed down the a-axis is shown in Figure 4.26. At first glance,
the structure may appear very similar to that of KCquuTe4. The layers are polar
and are composed of square antiprismatic europium atoms sandwiched between a
[AgTe]' anti-PbO type layer and a flat Te net, see Figure 4.27A. However, unlike
KCquuTea which has perfectly square Te net, The Te net in K0.65Ag2Eu1,35Te4 is
modulated, see Figure 4.28. The Te atoms have oligomerized into trimers and
heptamers which alternate across the layer in a 1:1 ratio. In addition to alternating
with each other, they also alternate back and forth in their direction. This
alternating pattern exists not only across the b-axis, but also across the c-axis.
Therefore, in order to re-desribe the periodicity of the structure, the b-axis must be
multiplied by 10 and the c-axis by 2. Conceptually, the unit cell of
K0,65Ag2Eu1,35Te4 can be considered as a law}, x 10b3,“, x 2cm supercell over that
of KCquuTe4. However, a direct comparison cannot be made because
KCquuTea possesses its own supercell of 1a,“), x 7bsub. The reason that the two
compounds possess different supercells and therefore different modulated Te nets
is because the charge per Te atom on the net is different for each (-0.5 for
KCquuTea and —0.675 for KogsAngumTea). This difference is due to the fact
that, in Ko_65Ag2Eu1_35Te4, the potassium Sites are disordered with europium (65%
K : 35% Eu). Because potassium is monovalent and europium is divalent, the

disorderin g of these elements affects the overall charge on the compound. These

212

-m_-_>t-.!.-.__-__,‘ ,. ._2x, ,_ ~. - .

disordered sites, much like the pure Eu sites, are 8—coordinate square antiprismatic
with Te, see Figure 4278. The fractional atomic coordinates, isotropic and
anisotropic temperature factors, bond distances and bond angles are given in
Tables 4.35-4.37. The three K/Eu disordered Sites were not refined anisotropically
due to their very small isotropic tempererature factors. While we do not yet fully
understand the origin of these small displacement parameters, we think that they
might be due to the fact that there was less than enough observed data. Much of
the crystallographic reflections that were expected to be present were either very
weak or absent. Overall, only 23% of the data were actually observed which
might explain these small displacement parameters. While we believe that this
structure model is a very close approximation to the true structure, it is possible
that there exists an additional modulation that is either incommensurate or too
weak to be observed by X-rays which currently we do not account for.
Considering that there still exists some disorder on the potassium Sites with

europium, this supercell could also succeed in resolving this disorder.

213

 

 

 

 

 

ellipsoids) viewed down the a-axis. The ellipses with octant Shading represent Eu
214

Figure 4.26 ORTEP representation of the structure of K0_65Ag2Eu1.35Te4 (80%
atoms, the crossed ellipses represent Ag atoms and the open ellipses represent Te

and K atoms.

 

F EK P01.) P”

 

L ‘-‘ ~_*.._'..- 9..."; . . x: o t.

(A)

Te7

Tel

 

TelOO O 0 Te12
‘ I’ ‘s I’ ‘ 6’
\ KIIBM ' mus , ' ‘s I’
I s‘ -.--.’ \‘KJIEN’I
Tel I... \ -....- t-...: O .-..—D .....~ ’ "-.- T012
,?‘ Te 1 0 3‘ Tel 1 I?‘
‘ ’ .' ‘ I U ‘
I :' s I .' s I :| s
I e I s I l I s I r I s
I l s I I s I 0 l s
” ' ‘s I’ O. ‘s I’ ‘s
Te2 , Tel : Te3 : Te5 : Te6
Te2 Te4 O
Tel Te3 Te5

Figure 4.27 ORTEP representation of (A) the coordination environment around
Eu in Ko.65Ag2Eu,,35Te4 and (B) the coordination environment around K/Eu in
Ko,65Ag2Eu1_35Te4 (90% ellipsoids for both). The ellipses with octant shading
rcpresent Eu atoms and the small and large open ellipses represent K/Eu and Te,
PCSpectively.

215

147’

Figure 4.28 ORTEP representation of the Te “net” of K0.6541821311135194 as seen
along the ab plane (80% probability ellipsoids) highlighting the arrangement of
tl‘imers and heptamers.

216

Table 4.35 Fractional Atomic Coordinates Isotropic

Displacement Parameters (Beg) for K0.65Ag2Eu. ,35Te4 (D() with Estimated Standard

Deviations in Parentheses.

and Equivalent

 

 

atom x y z occupancy Ueq‘, A2
Eu(1) 0.5070(13) 0.9500(1) 1.0527(2) 1.0 3.1(1)
Eu(2) 0.5054(13) 0.8500(1) 1.0526(1) 1.0 2.8(1)
Eu(3) 0.4887(14) 0.75 1.0526(1) 1.0 2.9(1)
Te(l) 0.50 1.00 0.9358(3) 1.0 24(1)
Te(2) 0.0297(7) 0.9497(1) 09365(2) 1.0 1.6(1)
Te(3) 0.4629(7) 0.9001(1) 0.9358(1) 1.0 1.3(1)
Te(4) 0.0149(9) 0.8501(1) 0.9367(2) 1.0 2.1(1)
Te(5) 0.5253(8) 0.7997(1) 0.9360(2) 1.0 19(1)
Te(6) 0.9614(9) 0.75 0.9367(2) 1.0 1.2(1)
Te(7) 1.00 1.00 1.1036(2) 1.0 1.2(1)
Te(8) 1.0057(6) 0.9000(1) 1.1036(1) 1.0 1.4(1)
Te(9) 0.0037(6) 0.7000(1) 1.1032(1) 1.0 1.3(1)
Te(lO) 0.5041(6) 0.9500(1) 1.2635(2) 1.0 2.0(1)
Te(l 1) 0.5010(6) 0.8500(1) 1.2630(2) 1.0 2.1(1)
Te(12) 0.4948(9) 0.75 1.2632(2) 1.0 2.0(1)
A20) 05 1.00 1.1841(3) 1.0 2.3(1)
A8(2) 1.0020(20) 0.9500(1) 1.1839(2) 1.0 2.3(1)
A8(3) 0.5130(20) 0.8999(1) 1.1845(2) 1.0 2.2(1)
A80) 0.9974(14) 0.8500(1) 1.1839(2) 1.0 26(2)
A8(5) 0.4910(16) 0.7000(1) 1.1847(2) 1.0 2.3(1)
A8(6) 0.0020(30) 0.75 1.1843(3) 1.0 2.6(1)
K0) 0.0 1.00 0.8168(2) 0.638(16) 0.3(2)
Eu(4) 0.0 1.00 0.8168(2) 0.362(16) 0.3(2)
K(2) 0.138(8) 0.9000(1) 0.8162(2) 095000) 01(1)
130(5) -0.l38(8) 0.9000(1) 0.8162(2) 0.35000) 0.2(1)
K(3) 1.0093(8) 0.7998(1) 0.8170(2) 0.656(11) 0.1(1)
1311(6) 1.0093(8) 0.7998(1) 0.8170(2) 03440 1) (“(1)

 

aUeq is defined as one third of the trace of the orthogonalized Uij

217

tCI‘ISOl'

 

Tc
Tc
It
It
Te

Table 4.36 Anisotropic Displacement Parameters (A) for K065Ag2Eu135Te4 (1X)
with Standard Deviations in Parentheses.

 

 

Atom U1 1 U22 U33 U23 U13 U12

1311(1) 0.027(1) 0.0035(1) 0.030(1) 0.0004(1) 0.0007(1) 0.0005(1)
Eu(2) 0.033(2) 0.0021(1) 0.030(1) 0.0003(2) 0.0000(2) 0.0002(1)
1311(3) 0.036(2) 0.0023(1) 0.028(2) 0 0.0003(2) 0

Te(l) 0.039(3) 0.0017(2) 0.015(3) 0 0 0.0015(2)
Te(2) 0.014(1) 0.0012(1) 0.021(2) 0.0001(1) 0.0004(2) 0.0000(1)
Te(3) 0.015(1) 0.0009(1) 0.016(1) 0.0001(1) 0.0006(1) 0.0001(1)
Te(4) 0.032(2) 0.0011(1) 0.018(2) 0.0001(1) 0.0001(2) 0.0008(1)
Te(5) 0.032(2) 0.0008(1) 0.017(2) 0.0001(1) 0.0006(1) 0.0006(1)
Te(6) 0.013(2) 0.0004(1) 0.020(2) 0 0.0002(2) 0

Te(7) 0.01 1(2) 0.0008(1) 0.018(2) 0 0 0.0002(1)
Te(8) 0.013(1) 0.0009(1) 0.019(1) 0.0000(1) 0.0001(1) 0.0001(1)
Te(9) 0.015(1) 0.0003(1) 0.022(2) 0.0001(1) 0.0005(1) 0.0001(1)
"16(10) 0.012(1) 0.0013(1) 0.034(2) 0.0000(1) 0.0001(1) 0.0001(1)
Te(ll) 0.014(1) 0.0013(1) 0.034(2) 0.0000(1) 0.0002(1) 0.0002(1)
Te(12) 0.016(2) 0.0006(1) 0.039(3) 0 0.0004(1) 0

Ag(1) 0.026(2) 0.0028(2) 0.015(3) 0 0 0.0003(3)
Ag(2) 0.027(2) 0.0030(2) 0.013(2) 0.0002(2) 0.0001(2) 0.0004(2)
Ag(3) 0.025(2) 0.0023(1) 0.018(2) 0.0002(2) 0.0001(2) 0.0004(2)
A8(4) 0.032(2) 0.0030(1) 0.017(2) 0.0003(2) 0.0005(2) 0.0004(3)
Ag(5) 0.026(2) 0.0017(1) 0.028(2) 0.0003(2) 0.0005(2) 0.0003(3)
Ag(6) 0.031(3) 0.0019(2) 0.029(3) 0 0.0002(3) 0
K1/Eu4 0.003(2)*

K2/Eu5 0.002(1)*

K3/Eu6 0.001(1)*

 

fie disordered K/Eu sites were isotropically refined only.

218

Table 4.37 Selected Distances (A) and Bond Angles (deg) for K0.65Ag2Eu1.35Te4

(IX) with Standard Deviations in Parentheses.

 

Bond Distances

Eul — Tel 3.496(6) Ag5 — Te12 2.886(5)

Eul — T62 3.411(6), 3.542(7) Ag6 — Te9 2.923(5)

Eul - Te3 3.501(5) Ag6 — T612 2.855(12), 2.905(12)
Eul — Te7 3.375(4), 3.416(4) Agl — Ag2 3.197(8), 3186(8)
Eul — Te8 3.393(5), 3.401(5) Ag2 — Ag3 3.159(14), 3.229(14)
Eu2 — Te3 3.505(4) Ag3 - Ag4 3.142(10), 3.239(11)
Eu2 — Te4 3.444(6), 3.499(7) Ag4 — Ag5 3.170(6), 3.210(6)
Eu2 -— Te5 3.501(4) Ag5 - Ag6 3.158(13), 3.227(14)
Eu2 - Te8 3.396(5) Tel — Te2 3.108(3)

Eu2 — Te9 3.366(5), 3.420(5) Te2 — Te3 2.974(4)

E113 — Te5 3.487(5) Te3 — Te4 3.033(5)

E113 — Te6 3.3552(8), 3.393(8) Te4 - T65 3~170(5)

Eu3 — Te9 3.371(5), 3.416(5) Te5 — Te6 2986(4)

Agl — Te7 2.902(5) K1/Eu4 - Tel 3.525(6)

Agl — Te10 2.899(5) Kl/Eu4 -— Te2 3.556(5)

Ag2 — Te7 2.91 1(5) Kl/Eu4 — Te10 3.404(3), 3.428(3)
Ag2 — T68 2.912(5) K2/Eu5—Te2 3.555(5)

Ag2 — Te10 2.882(10), 2.898(9) K2/Eu5 - Te3 3463(5), 3-602(5)
Ag3 — T68 2.885(9), 2.934(9) K2/Eu5—Te4 3.558(5)

Ag3 _ Te10 2.895(5) K2/Eu5 — T610 3.383(4), 3.441(4)
Ag3 _ Tell 2.335(5) K2/Eu5 — Tell 3.377(4), 3.453(4)
Ag4 _ Te8 2.913(5) K3/Eu6 — Te4 3.554(5)

Ag4 _ T39 2.916(5) K3/Eu6 - Te5 3.480(5), 3.571 (5)
Ag4 — Tel 1 2.870(7), 2.895(7) K3/Eu6 - Te6 3.545(5)

A85 _ Te9 2.899(7), 2.936(8) K3/Eu6 — Tell 3.397(4), 3.457(4)
Ag5 _ Tell 2.880(5) K3/Eu6 — Te12 3.400(5), 3.424(5)

 

 

219

 

Te
Te
Te
1e

Table 4.37 continued

Selected Distances (A) and Bond Angles (deg) for

Ko,65Ag2Eu1,35Te4 (IX) with Standard Deviations in Parentheses.

 

 

Bond Angles

Tel - Eul - Te2 5347(8), 5583(9) Te6 — Eu3 — Te9 81.03(16), 132.72(10)
Tel — Eul — Te3 80.61(11) Te9 — Eu3 — Te9 8303(7), 140.25(17)
Tel - Eul — Te7 79.89(10), 80.46(11) Te2 — Tel — Te2 179.4(2)

Tel — Eul - Te8 133.25(17), 134.33(17) Tel — Te2 — Te3 96.14(10)

Te2 — Eul — Te2 80.62(1 1) Te2 — Te3 - Te4 97.40(12)

Te2 - Eul — Te3 5096(8), 57.69(9) Te3 — Te4 — Te5 177.55(16)

Te2 — Eul — Te7 79.85(15), 134.43(1 l) Te4 — Te5 - Te6 94.90(15)

Te2 - Eul — Te8 79.30(14), 133.8500) Te5 - Te6 — Te5 97.83(18)

Te3 — Eul — Te7 130.44(17), l37.08(17) Te7 — Agl - Te7 101.6(2)

Te3 - Eu 1 — Te8 78.00(10), 82.34(10) Te7 - Agl — Te10 1 1293(6), l 1354(6)
Te7 - Eul — Te7 8298(7) Te7 - Ag2 - Te8 102.10(17)

Te7 - Eul — Te8 83.25(13), 140.1303) Te7 - Ag2 — Te10 113.2(3)

Te8 — Eul - Te8 8249(9) Te8 - Ag2 — Te10 112.9(3), 113.6(3)
Te3 — Eu2 - Te4 5195(9), 5737(9) Te10 — Ag2 — Te10 102.21(18)
Te3—Eu2—Te5 81.11(9) Te8—Ag3—Te8 101.28(17)

Te3 — Eu2 — Te8 78.00(10), 82.2100) Te8 — Ag3 — Te10 112.3(2), 114.0(2)
Te3 — Eu2 - Te9 130.41(17), l37.02(17) Te8 — Ag3 — Tell 112.1(3), 114.3(2)
Te4 — Eu2 — Te4 80.78(l l) Te10 — Ag3 - Tell 103.19(16)

Te4 — Eu2 - Te5 53.8600), 5553(9) Te8 - Ag4 — Te9 101.94(16)

Te4 - Eu2 — Te8 80.66(15), 133.35(10) Te8 — Ag4 — Tell 112.43(19), 113.89(18)
Te4 - Eu2 — Te9 80.20(15), 133.17(9) Te9 — Ag4 — Tell 113.20(19)

Te5 — Eu2 - Te8 132.5508), 135.2608) Tell — Ag4 - Tell 102.5708)

Te5 - Eu2 — Te9 78.8200), 80.9600) Te9 - Ag5 — Te9 100.880 7)

Te8 - Eu2 — Te8 8294(8) Te9 -— Ag5 — Tel 1 1 12.4(2), 1 14.2(2)
Te8 — Eu2 — Te9 84.0503), 140.2901) Te9 — AgS - T912 1128(2). 113-7(2)
Te9 — Eu2 - T69 3105(7) Tell — Ag5 — Te12 103.4008)

T65 — Eu3 — Te5 8039(15) Te9 — A86 - Te9 101-5(2)

Te5 — Eu3 — Te6 51.4100), 57.5901) Te9 — A86 - T912 112-“3411390)
Te5 - Eu3 — Te9 8186(9), 136.5(2) Te12 — Ag6 - T912 102-7(3)
Te6-Eu3-Te6 80.7105)

 

220

Table 4.37 continued

Selected Distances (A) and Bond Angles (deg) for

K0.65AnguL35Te4 (IX) with Standard Deviations in Parentheses.

Bond Angles
Tel — Kl/Eu4 — Tel

Tel - K1/Eu4 — Te2
Te2 - Kl/Eu4 —- Te2
Tel — Kl/Eu4 — Te10
Te2 -— Kl/Eu4— Te10
Te10 — Kl/Eu4 - Te10
Te2 — K2/Eu5 — Te3
Te2 — K2/Eu5 — Te4
Te2 - K2/Eu5 — Te10
Te2 — K2/Eu5 — Tel 1
Te3 - K2/Eu5 — Te3
Te3 — K2/Eu5 - Te4
Te3 - K2/Eu5 - Te10
Te3 — K2/Eu5 — Tell
Te4 — K2/Eu5 - Te10
Te4 - K2/Eu5 - Tel 1
Te10 — K2/Eu5 — Te10
Te10 — K2/Eu5 — Tell
Tell - K2/Eu5 — Tell
Te4 - K3/Eu6 - T65
Te4 - K3/Eu6 — Te6
Te4 — K3/Eu6 - Tel 1
Te4 — K3/Eu6 —- Te12
Te5 - K3/Eu6 —- Te5
Te5 - K3/Eu6 — Te6
Te5 — K3/Eu6 - Tel 1
Te5 - K3/Eu6 - Te12
Te6 — K3/Eu6 — Tel 1
Te6 - K3/Eu6 - Tel 2
Tell — K3/Eu6 -— Tell
3L1 - K3/Eu6 — Te12

79.3207), 5583(9)

5207(9), 5547(9)

79.7604)

8142(7), 81.750), 133.8900), 134.0500)
7984(8), 8277(8), 132.0801), 136.4301)
8238(8), 8306(9), 38.4609)

5009(9), 56.69(9)

78.7700)

7937(9), 83.4300)

130.56(12), 137.5703)

7902(11)

51.13(9),56.01(10)

80.8800), 8202(10), 132.9402), 134.3401)
81.3200), 82.2100), 133.2502), 134.7601)
131.5803), 136.3204)

80.2500), 82.83(11)

8250(8)

8251500), 1384503)

8239(9)

53.5700), 5402(10)

79.4500)

81.0905), 81.7800)

134.1905), 134.5405)

79.50 1)

5029(9), 5692(10)

81.6300), 8212(10), 133.580), 134.6601)
81.2102), 82.21(12), 133.7502),134.61(11)
130.6703), 138.0504)

79.4701), 84.050 1)

8206(9)
8225(11), 83.5001), 137.7304), 137.9704)

221

Magnetic Susceptibility and Infiared Spectroscopy of K0. 6 5Ag2Eu ,_ 35T e4 (1)0
Variable temperature magnetic susceptibility data for K0_65Ag2Eu.,35Te4 were
measured over the range of 5-300K at 30006. A plot of l/xM vs T (see Figure
4.29) shows that this material exhibits perfect Curie-Weiss behavior. A par value
of 8.53 BM and a Weiss constant of -—231K was estimated by fitting a straight line

to the data above 50K. These values are consistent with an 1‘7 configuration for
Eu2+ (7.9 - 8.0 BM) and are very different from that expected for EU“ (3.3-3.5
BM)”

The diffuse reflectance optical spectra was taken in the Mid-IR
region for K065Ag2EuL35Te4, see Figure 4.29B. An abrupt optical gap is observed
at 0.28eV, which can be assigned as the bandgap and therefore the material is a
semiconductor. Below 0.2eV, there is another optical transition, possibly due to f-

f interactions on the europium center.

222

 

l/’Y

 

35irtrIrII—UI—i—"IIIIIIIIWYTrrIIrr

 

O
,0 (A) . .
. .
25 e
2 O
X 20 . .
\
"‘ 15 , '
O
C
10 e.
..C
5 a".
O ILILILILIUIJI+1LILIIJIkill._
0 50 100 150 200 250 300
Temperature(K)

(B)

 

1.
A 1
g 1.
g . .9 ‘3
E .4341?" 1
In 4
5; 1
{J J
a 1
11

0.2 0.3 0.4 0.5 0.6 0 7 0.8

Energy (eV)
Figure 4.29 (A) Inverse molar magnetic susceptibility (l/xM) plotted against

temperature (2-300K) for KwsAngumsTe... (B) Diffuse reflectance optical
SPeCtra (in the Mid-IR region) for K0_65Ag2Eu1.35Te4.

223

Charge Transport Measurements of K0,65Ag2Eu1.35Te4 - Electrical
conductivity and thermoelectric power data were measured as a function of
temperature for four single crystals of K0,65Ag2Eu1_35Te4, see Figure 4.30. The
conductivity data for all crystals suggest semiconducting behavior with a sharp
decline below 150K. However, the magnitudes differ significantly from crystal to
crystal. The room temperature values for these four crystals were 29 S/cm, 132
S/cm, 692 S/cm, and 2892 S/cm. This wide range is very unusual and suggests a
problem with either the measurement or the sample. Therefore, measurements
were made on several more crystals and the results showed room temperature
conductivity values ranging from 10-100 S/cm. While this range is still somewhat
large, it gives a better idea as to the conductivity on average. Although the origin
of this varying conductivity is still not understood, the fact that it was reproduced
suggests that it is particular to the compound and not a problem with the
measurement.

The thermopower data for two of the four initial crystals are shown in
Figure 4.30B. Although the conductivity values exhibited a wide range, the
thermopower values were consistent with a very narrow range of 170-190 uV/K at
room temperature. The positive sign and decreasing Seebeck coefficient with

falling temperature is consistent with a semiconductor and suggest p-type

behavior.

224

 

 

 

 

 

 

 

 

(A)
103
”a“
Q 102 O o N
{’3 /
b 1 .
00 10
.3 0’
I/°
100 ° °
.f
10'1 2.._rlm..nl._s..1..4.14..21.g.1...1
50 100 150 200 250 300 350
Temperature (K)
200 ThermoEMer
E(B) 4?
A 150 t .3"; I 1
1 2 '
g ' ~ 1
)- I a
V ' .0 j
W F .r 1
so 1—.‘ ..o“. 1
I so ' 1
1, o
o .- 4 A A l . . . - l . A ALI -4¥L A4. A‘.+L4+.gl
0 50 100 150 200 250 300

Temperature (K)

Figure 4.30 (A) Four probe electrical conductivity data for single crystals of
Ko_65Ag2Eu1,35Te4 as a filnction of temperature. (B) Thermopower data for single

crystals of K0.65Ag2Eu1,35Te4 as a fimction of temperature.

225

D. Conclusions

Several new compounds of the type AWManyTez have been discovered
through the use of alkali metal/polytelluride fluxes. The common theme that runs
through these compounds is the existence of a Te net. Futhennore, these Te nets
have been found to be, in most cases, distorted with the actual distortion being
highly dependent on the average charge per Te atom in the net. Below is a
summary table which illustrates how very small changes in the average charge can

have a drastic effect on the resulting modulation.

 

 

 

 

 

 

Compound Name Average charge per Supercell observed
Te atom in the net
KCuCeTe4 -0.50 lasub X 2-87bsub
(or 2.87am x 2.87bsub)
K2,5Ag4,5Ln2Te9 -0.75 lasub X 3bsub
(Ln = Ce, La)

KCU2EUTC4 -0.5 1asub X 7bsub

I¥I(0.65Ag2EU1.35TC4 -O.675 lasub X 10bsub X 2csub

 

 

 

Interestingly, KCuCeTe4 and KCquuTe4 possess the same average charge but
exhibit different supercells. This is an indication that the modulation in the net not
only is affected by the electron count, but also by the remaining part of the
structure and its makeup. While some of these Te net distortions were able to be
elucidated, others were beyond our capabilities. However, extensive collaboration

has been initiated to try and solve the extraordinarily weak or incommensurate

226

 

supercells. Most of these materials were determined to be p-type semiconductors
with narrow bandgaps ranging around 0.2-0.3 eV. The semiconducting behavior
is understood to be caused by the modulations that exist in the Te nets since they
act to lower the energy levels at the Fermi level and open up a gap. However, to
fillly understand the nature of these Te net distortions, further experimental and
theoretical work is needed. More of these types of compounds need to be studied
in order to systematically build a relationship between the electron count and the

type of distortion.

227

References

 

Sutorik, A.C.; Albritton-Thomas, J.; Kannewurf, C.R.; Kanatzidis, M.G. J.
Am. Chem. Soc. 1994, 116, 7706.

Sutorik, A.C.; Albritton-Thomas, J.; Hogan, T.; Kannewurf, C.R.;
Kanatzidis, M.G. Chem. Mater. 1996, 8, 751.

Wu. P.; Ibers, J.A. J. Solid State Chem. 1994, 110, 156.
Cody, J.A.; Ibers, J.A. Inorg. Chem. 1995, 34, 3165.

Cody, J.A.; Mansuetto, M.F.; Pell, M.A.; Chien, S.; Ibers, J.A. J. Alloys
Comp. 1995, 219, 59.

Pell, M.A.; Ibers, J .A. Chem. Ber./Recueil 1997, 130, l.

(a) Christuk, A.E.; Wu, P.; Ibers, J.A. J. Solid State Chem. 1994,110, 330.
(b) Wu. P.; Ibers, J.A. J. Solid State Chem. 1994, 110, 337.

Huang, F .-Q.; Choc, W.; Lee, S.; Chu, J. Chem. Mater. 1998, 10, 1320.

Patschke, R.; Heising, J .; Schindler, J.; Kannewurf, C.R.; Kanatzidis, M.G.
J. Solid State Chem. 1998, 135, 111.

Zhang, X.; Schindler, J.L.; Hogan, T.; Albritton-Thomas, J.; Kannewurf,
C.R.; Kanatzidis, M.G. Angew. Chem. Int. Ed Engl. 1995, 32, No. 1, 68.

Zhang, X.; Li, J.; Foran, B.; Lee, S.; Guo, H.-Y.;Hogan, T.; Kannewurf,
C.R.; Kanatzidis, M.G. J. Am. Chem. Soc. 1995, 117, 10513.

Gweon, G.H.; Denlinger, J.D.; Clack, J.A.; Allen, J.W.; Olson, C.G.;
DiMasi, E.; Aronson, M.C.; Foran, B.; Lee, S. Phys. Rev. Lett. 1998, 81,
886.

Patschke, R.; Heising, J.; Brazis, P.; Kannewurf, CR; Kanatzidis, M.G.
Chem. Mater. 1998, 10. 695.

Norling, B.K.; Steinfink, H. Inorg. Chem. 1966, 5, 1488.

Kronert, V.W.; Plieth, K.Z. Z. Anorg. Allg. Chem. 1965, 336, 207.

228

. - - l- u L fit..-

 

16

l8

19

20

2]

22

23

24

25

26

27

28

29

(a) Greenwood, N.N.; Eamshaw, E Chemistry of the Elements; Pergamon
Press: New York, 1984; 1434. 0)) Guittard, M.; Flahaut, J. Synthesis of
Lanthanide and Actinide Compounds, Kluwer Academic Publishers,
Netherlands, 1991, 321.

Bensch, W.; Dilrichen, P. Chem. Ber. 1996, 129, 1489.

Patschke, R.; Brazis, P.; Kannewurf, C.R.; Kanatzidis, MG J. Mater.
Chem. 1999, 9, 2293.

Feher, F. Handbuch der Praparativen Anorganischen Chemie: Brauer, G.,
Ed.; Ferdinand Enke: Stuttgart, Germany 1954, 280.

Sheldrick, G.M., in Crystallographic Computing 3; Sheldrick, G.M.;
Kruger, C.; Doddard, R., Eds.; Oxford University Press: Oxford, England
1985, 175.

Gilmore, G.J., Appl. Cryst. 1984, 17, 42.

SMART: Siemens Analytical Xray Systems, Inc., Madison, WI, 1994.

SAINT: Version 4.0, Siemens Analytical Xray Systems, Inc., Madison WI,
1994-1996.

SADABS: Sheldrick, GM. University of thtengen, Germany, to be
published.

Sheldrick, GM. SHELXT L, Version 5; Siemens Analytical Xray Systems,
Inc.; Madison, WI, 1994.

(a) Lin, W.; Steinfink, H.; Weiss, F. J. Inorg. Chem. 1965, 4, 877. (b)
Wang, R.; Steinfink, H.; Bradley, W.F. Inorg. Chem. 1966, 5, 142. (c)
Pardo, M.-P.; F lahaut, J.; Domange, L.C.R. Bull. Soc. Chim. Fr. 1964,
3267. ((1) Ramsey, T.H.; Steinfink, H.; Weiss, E. J. Inorg. Chem. 1965, 4,
1154. (e) Norling, B.K.; Steinfink, H. Inorg. Chem. 1966, 5, 1488.

Cody, J.; Ibers, J.A. Inorg. Chem, 1996, 35, 3836.

Wachhold, M.; Sheldrick, W.J. Angew. Chem. 1995, 107, 490.; Angew.
Chem. Int. Ed Engl. 1995, 34, 2109.

Lee, S.; Foran, B. J. Am. Chem. Soc. 1994, 116, 154-161.

229

.____.——--—r-

*A 'h -_ M ‘ " __._‘ w»— ‘1‘_.——*2 A: . 4.44 .- -

 

30

31

32

33

34

35

36

37

38

Park, Y.; Kanatzidis, M.G., unpublished results.

(a) Savelsberg, G.; Schafer, H. Z. Naturforsch 1978, 33b, 370-373. (b)
Berger, R.; Eriksson, L. .1. Less Common Met. 1990, 161, 101.

Greenwood, N. N.; Eamshaw, A. Chemistry of the Elements; Pergamon
Press: New York, 1984; p 1443.

Brown, D.B.; Zubieta, J.A.; Vella, P.A.; Wrobleski, J.T.; Watt, T.; Hatfield,
W.B.; Day, P. Inorg. Chem. 1980, 19, 1945.

(a) Anton, H. Elementary Linear Algebra, 7‘11 edition, Chapter 3, John
Wiley & Sons, New York. (b) Megaw, H.D. Crystal Structures: A Working
Approach, Chapter 10, 1973, W.B. Saunders Company, Philadelphia,
London, Toronto.

Cordier, G.; Schafer, H.; Woll, P. Z. Naturforsch. 1985, 40b, 1097.

(a) Gweon, G.-H.; Denlinger, J.D.; Clack, J.A.; Allen, J .W.; Olson, C.G.;
DiMasi, E.; Aronson, M.C.; Foran, B.; Lee, S. Phys. Rev. Lett. 1998, 81,
886. (b) DiMasi, E.; Foran, B.; Aronson, M.C.; Lee, S. Phys Rev. B. 1996,
54, 13587. (c) DiMasi, E.; Aronson, M.C.; Mansfield, J.F.; Foran, B.; Lee,
S. Phys. Rev. B. 1995, 52, 14516. (d) Foran, B.; Lee, S. J. Am. Chem. Soc.
1994, 116, 154. (e) DiMasi, E.; Foran, B.; Aronson, M.C.; Lee, S. Chem.
Mater. 1994, 6, 1867. (e) Foran, B.; Aronson, M.C.; Lee, S. Anderson,
M.C. Chem. Mater. 1993, 5, 974.

Zhang, X.; Li, J.; Foran, B.; Lee, S.; Guo, H.-Y.; Hogan, T.; Kannewurf,
C.R.; Kanatzidis, M.G. J. Am. Chem. Soc. 1995, 117, 10513.

Hanko, J .A.; Kanatzidis, M.G.; Evain, M.; Gourdon, O.; Boucher, F.;
Submitted for publication.

230

we 0" r. - ' a ‘. .__._._-..~ ifl- -u“-—-.-o-——-' -«--—--—

 

Chapter 5

Novel Quaternary Polytelluride Compounds Without Te Nets

231

—-—-h‘-—-L “n04” '1“:

A. Introduction

Over the past decade, the polychalcogenide flux method has become an
established technique for discovering new solid state chalcogenides.l Although
many of the compounds discovered by this method form completely new structure
types, others are reminiscent of and can be considered derivatives of known
chalcogenides. This is particularly true when lanthanide and actinide metals are
involved. The binary LnQ3 phases (NdTe32 and ZrSe33-type), for example, are
quite stable. In fact, several new ternary phases have recently been reported in
which the structural motifs are related to these LnQ3 binaries. While NaLnS3
(Ln=La.,Ce)4 and ATh2Q6 (A = Cs, Rb, K; Q = Se, Te)5 represent two different
variations of the ZrSe; structure type, ALn3Te3 (A = Cs, Rb, K; Ln = Ce, Nd)6 is
closely related to the structure of NdTe3. In an effort to access quaternary phases
which are less structurally related to the LnQ; binaries, another element was
introduced into the synthesis. Copper and silver have proven to be particularly
well behaved in this respect and we were able to isolate several new compounds,
whereas other elements gave phase-separated ternary compounds. Prior

investigations into the A/Cu/Ln/Q (Q=S, Se) system produced several quaternary
7,8
compounds, including KzCuzCeS4,7 KCuCeZS6, , KCuLa286,8 CsCuCe286,8
8
KCuCeZSe6,8 CsCuCeS3,8 and KCuUSe3 ~ Of these. I<2Cu2CeS.7 and (380109838

exhibit mixed chalcogenide valency and appreciable electrical conductivity. At

the same time, there has been a rapid expansion in this area by independent

232

__—~.._____ ___.. _..

investigators producing such compounds as CsCuUTe3,9 BaLnMQ3 (Ln=La, Ce,
Nd, Er; M=Cu,Ag; Q=s,8e),lo BaDyCuTe3,” KUDyzCustes,“
K05Ba05DyCuL5Te3,” and KCuEu286'2. These results support the premise that by
combining the ionic lanthanide and actinide bonding with the more covalent
transition metal bonding, one can access phases with novel structures and
properties. I“ Furthermore, it has become increasingly apparent that the greater the
amount of copper or silver in the framework, the more profound the effect of
breaking up the known structure types. Of course, a better understanding of this
newly emerging family of compounds could be achieved if a wider variety of
members were available for study, including the corresponding tellurides.
Because of this, we decided to examine the A/M/Ln/T e (M=Cu,Ag) system using
polytelluride fluxes and as a result discovered two new quaternary phases,
K2Ag3CeTe4l3 and szCu3CeTe5'4. KzAg3CeTe4 has a three-dimensional tunnel
framework built from the linking together of [AgZCeTe4]3' layers with Ag and
displays the rare combination of being a narrow-gap semiconductor and at the
same time being accessible through ion-exchange, while szCu3CeTe5 is two-
dimensional and is a perfect example of how the basic LnQ3 framework can be

substantially broken up to form a higher order phase.

B. Experimental Section

1. Reagents — The following reagents were used as obtained: Potassium

metal, analytical reagent, Spectrum Chemical Mfg. Corp., Gardena, CA;

233

mu‘r A‘L-

Rubidium metal, 99.5%, Alfa Aesar, Ward Hill, MA.; Copper metal, electrolytic
dust, Fisher Scientific, Fairlawn, NJ; Silver metal coin, 99.9% purity, Liberty
Coin, Lansing, MI; Cerium metal, < 250 mesh, Alfa Aesar, Ward Hill, MA;
Tellurium powder, 100 mesh, 99.95% purity, Aldrich Chemical Co., Milwaikee,
WI; N, N, - Dimethylforrnamide (DMF) was used as obtained in analytical
reagent grade fi'om Aldrich Chemical Co., 998% purity, Milwaukee, WI.

Silver Powder — A silver coin weighing 31.54g was dissolved in 250 mL
of 8.4M HNO3. The solution was heated to 60°C in an acid-resistant fume hood
until the silver coin was completely dissolved. The solution was neutralized with
ammonium hydroxide and the silver was reduced with formic acid until a pH of 7-
8 was reached. The resulting pale grey solid was filtered, washed with copious
amounts of distilled water and acetone, and dried in vacuum overnight. The final
yield was 31.095g.

Potassium T elluride, K 2T e —— Synthesis of this material was performed as
described in Chapter 2, Section B.1

Rubidium Telluride, szT e -— Synthesis of this material was performed as

described in Chapter 2, Section B]

2. Synthesis — All manipulations were carried out under a dry nitrogen
atmosphere in a Vacuum Atmospheres Dri-Lab glovebox.
KzAg3CeTe4 (I) — Amount of 0.309g KzTe (1.5 mmol), 0.162g Ag (1.5

mmol), 0.070g Ce (0.5 mmol), and 0.447g Te (3.5 mmol) were weighed into a vial

234

.,_ _. . ._ .._,.—,‘,_ __.

in an N; filled glovebox. The reagents were thoroughly mixed and loaded into a
9mm carbon coated silica ampoule. The ampoule was removed from the
glovebox, evacuated ‘on a Schlenck line to less than 2.0 x 10‘4 mbar, and flame
sealed. The reactants were heated to 400°C in 12hrs, isothermed at that

temperature for 12 hrs, raised to 850°C in 22hrs, and isothermed at that

temperature for 6 days. The tube was then cooled to 400°C at -4°C hr}, and then

quenched to room temperature in 4 hrs. The excess KxTey flux was removed,
under N2 atmosphere, with DMF to reveal black needle-shaped crystals which

appeared to be both air and water stable. Typical yields were 57% yield, based on
Ag. Phase homogeneity was confirmed by comparing the powder X-ray
diffraction pattern of the product against one calculated using the
crystallographically determined atomic coordinates, see Table 5 .1
Semiquantitative microprobe analysis on single crystals gave an average
composition of KLgAngCCTClb.

szCu3CeTe5 (II) - Amounts of 0.448g szTe (3.0 mmol), 0.095g Cu
(3.0 mmol), 0.070g Ce (1.0 mmol), and 0.447g Te (7.0 mmol) were weighed into a
vial in an N2 filled glovebox. The reagents were thoroughly mixed and loaded
into a 9mm carbon coated silica tube. The ampoule was removed fiom the
glovebox, evacuated on a Schlenck line to less than 2.0 x 104 mbar, and flame
sealed. The reactants were heated to 850°C for 24hrs, isothermed at that

temperature for 10 days, cooled to 400°C at —3°C/hr, and further cooled to room

235

temperature at -10°C/hr. The excess beTey flux was removed, under nitrogen
atmosphere, with dimethylformamide to reveal black needle-shaped crystals which
appeared to be both air and water stable. Typical yields were 45% yield (based on
Cu). Phase homogeneity was confirmed by comparing the power X-ray
diffraction pattern of the product against the one calculated using the
crystallographically determined atomic coordinates, see Table 5.2.
Semiquantitative microprobe analysis carried out on randomly selected crystals

gave an average composition of Rb2,5Cu3,3CeLoTe5_6.

236

 

 

Table 5.1 Calculated and Observed X-ray Powder Diffraction Pattern of
K2Ag3CeTe4 (I)
h kl dc“, (A) dobs(A) l/l,m (obs) (%)
l 0 1 11.4751 12.1064 23.4
2 0 0 8.5992 8.5191 100.0
0 0 2 7.7027 7.6691 49.2
2 0 2 5.7375 5.7863 1.8
3 0 1 5.3729 5.3997 6.4
1 0 3 4.9205 4.9232 2.7
2 0 3 4.4089 4.4030 20.7
2 1 0 4.0830 4.0728 6.5
0 0 4 3.8514 3.8395 13.6
4 0 2 3.7543 3.7376 18.5
2 l 2 3.6075 3.6448 5.0
4 0 3 3.2966 3.2794 12.9
201 3.1959 3.1800 11.7
5 0 2 3.1408 3.1276 12.9
4 l 1 3.0895 3.0743 58.1
2 0 5 2.9005 2.8974 51.4
2 1 4 2.8016 2.7878 8.6
3 0 5 2.7140 2.6879 7.4
l 0 6 2.5394 2.5488 13.9
6 0 3 2.5029 2.5249 14.4
2 0 6 2.4603 2.4876 4.2

 

237

Table 5.1 continued

Calculated and Observed X-ray Powder Diffraction

 

 

Pattern of KzAg3CCTC4 (I)
'1 kl dnnlfi) 4.1.14) III... (obs) 1%)
7 0 1 2.4263 2.4234 25.7
3 l 5 2.3426 2.3579 12.0
020 2.3196 2.3131 91.7
4 1 5 2.2038 2.1912 72.3
8 0 0 2.1498 2.1413 20.6
8 01 2.1293 2.1186 17.1
4 2 0 2.0415 2.0432 26.0
4 2 2 1.9734 1.9722 17.2
5 2 2 1.8659 1.8688 5.7
7 1 5 1.7748 1.7743 20.4
6 2 2 1.7557 1.7567 15.5
10 0 1 1.7092 1.7086 4.4
2 2 6 1.6878 1.6885 8.4
9 0 5 1.6240 1.6219 14.4
4 0 9 1.5903 1.5910 17.5
8 2 1 1.5686 1.5653 17.0
2 3 0 1.5220 1.5231 46.8

 

238

Table 5.2 Calculated and Observed X-ray Powder Diffiaction Pattern for
szCU3CCT65 (ID

 

 

1.1.1 and) 45(4) 111.... (obs) 1%)
0 0 1 11.4616 11.5477 100.00
2 0—1 8.8586 8.7086 49.46
2 0 0 8.3862 8.6142 15.43
2 0—2 5.9379 5.9876 5.98
0 0 2 5.7621 5.7738 22.66
4 0-1 4.6595 4.6714 10.16
4 0 0 4.3223 4.3071 7.33
2 0-3 4.1504 4.1667 6.88
0 0 -3 3.8482 3.8492 19.31
3 1-3 3.3925 3.3622 23.51
0 2 0 3.1139 3.1990 23.35
6 0-1 30810 3.0800 25.54
4 0 2 2.9578 2.9481 20.94
0 0 4 2.8893 2.8869 32.42
0 2 2 2.7405 2.7442 14.51
5 1 1 2.6983 2.7027 10.92
4 2 0 2.5272 2.5262 8.67
4 0—5 2.4796 2.4732 14.77
0 2 3 2.4219 2.4233 8.13
4 2 1 2.3559 2.3585 9.76

 

239

Table 5.2 continued Calculated and Observed X—ray Powder Diffraction

 

Pattern for szCu3CeTe5 (H)

h kl 9....00 95(4) III... (obs) 1%)
8 0 —2 2.3359 2.3357 9.17
0 O 5 2.3133 2.3095 11.13
1 1—5 2.2541 2.2526 12.76
6 2—2 2.1944 2.1961 10.37
8 0 0 2.1480 2.1536 18.37
6 2 0 2.1151 2.1125 9.97
4 0—6 2.0777 2.0833 9.43
1 3—1 2.0502 2.0487 9.38
6 2—4 2.0032 1.9976 15.22
0 0 6 1.9285 1.9246 17.23
10 0 0 1.7234 1.7228 10.00
8 0—7 1.6536 1.6553 31.55
0 0 7 1.6497 1.6497 19.92

12 0 -4 1.5461 1.5464 6.47

 

240

3. Physical Measurements - The instrumentation and experimental setup
for the following measurements are the same as described in Chapter 2, Section 3:
Semiquantitative Energy Dispersive Spectroscopy (EDS), Powder X-ray
Diffraction, Magnetic Susceptibility Measurements, and Charge Transport
Measurements. The instrumentation and experimental setup for the Infrared
Spectroscopy measurements is the same as described in Chapter 3, Section 3.
X.ray Crystallography - The single crystal data sets of both
KzAg3CeTe4 and szCu3CeTe5 were collected at the University of Minnesota by
Dr. Victor G. Young, Jr. A single crystal of each was mounted on the tip of a
glass fiber. Intensity data were collected at 293K (for K2Ag3CeTe4) and 173K (for
szCu3CeTe5) on a Siemens SMART Platform CCD diffiactometer using graphite
monochromatized Mo K01 radiation. The data were collected over a hemisphere of
reciprocal space, up to 50° in 20. The individual frames were measured with an 0)
rotation of 03° and an acquisition time of 60 sec/flame for K2Ag3CeTe4 and 30
sec/frame for szCu3CcTe5. The SMART15 software was used for the data
acquisitions and SAINT ‘6 for the data extractions and reductions. The absorption
corrections were performed using SADAB S. '7 The structures were solved by
direct methods using the SHELXTL18 package of crystallographic programs. The
complete data collection parameters and details of the structure solutions and

refinements are given in Table 3.3.

241

WHY-s

Inductively Coupled Plasma Spectroscopy (ICP) - Samples were
submitted to the Animal Health Diagnostics Laboratory at Michigan State
University for analysis.19 Experiments were carried out on a Thermo J arrell Ash
Polyscan 61E Simultaneous/ Sequential inductively coupled plasma-atomic
emission spectrometer (ICP-AES) with vacuum spectrometers and Ar-purged
optical paths. The solid powders were weighed onto an analytical balance and
then digested in a teflon container in concentrated nitric acid overnight at 95°C.
The digest was transferred to a 25mL volumetric flask and diluted with water.
Yttrium was used as an internal standard in 2% HNO3. Multielemental analyses
were done by nebulizing the liquid sample into an argon flame (plasma) that was
sustained by a surrounding high frequency magnetic field. The photons emitted
by the diffracting grating are collimated and directed by a diffraction grating onto
a semicircular array of photomultiplier tubes, one for each element to be
measured. A computer then converts the photomultiplier signals to concentration

units.

242

__ _.__.._._ ...I... AL M-A' .-

u «~fi—m.— u... _—.._-._.-_~- ._ ._

 

 

Table 5.3 Crystallographic Data for K2Ag3CeTe4 (I) and szCu3CeTe5 (H)
Formula KzAg3CeTe4 szCu3CeTe5
a, (A) 17.19850) 18.68840)
b, (A) 4.6393(2) 6.2384(2)
c, (A) 15.4055(3) 12.52640)
13, (deg) 90.000(0) 112.7950)
v, (A ) . 1229.190) 1346.34(5)
Space Group ana (#62) C2/m (#12)
Z value 4 4
F.W (g/mol) 526.16 1526.63
dd, (g/cm3) 5.686 5.623
p, (cm!) 18.262 25.741
crystal (mm3) 0.31x0.02x0.01 0.16x0.04x0.01
Radiation Mo K01 Mo K01
20...... (deg) 50.0 50.0
Temp., (°C) 293 173
No. data collected 5536 3427
No. unique data 1221 1307
R(int) 0.030 0.044
No. F02>20 (F02) 1066 1087
No. variables 62 6O
Rl/wR2, % . 3.2/6.6 4.6/11.8
Goof 1.06 1.04

 

a 2 r2 2r
Rl=2(|F0|-|Fcl)/2|Fol wR2={XIW(IFo l-IFc ll/ZWIFOI}

243

.._ —-—-r-—~_.

C. Results and Discussion

Structure Description of K 2Ag3CeT e4 (1) - The structure of KZAgBCeTe4 is
somewhat related to that of KzCuzCeS4,7 see Figure 5.1. The basic units that make
up the anionic framework in both compounds are [CeQé] octahedra and [MQ4]
(M=Cu,Ag) tetrahedra. In K2C112CeS4, layers are formed when double rows of
edge sharing [CuS4] tetrahedra alternate with chains of [CeS6] octahedra (Figure
5.2A). The layer in K2Ag3CeTe4, however, is now corrugated due to the different
way in which the chains of [CeTe6] octahedra connect to the double rows of
[AgTe4] tetrahedra (Figure 5.2B). In KzCuZCeS4 the rows of CuS4 tetrahedra are

arranged centrosymmetrically around chains of CeS6 octahedra, while in
K2Ag3CeTe4 the edge-sharing with [AgTe4] tetrahedra involves adjacent edges of
[CeTe6] octahedra. This difference creates a quadruply bridging Te atom (binding
to two Ag and two Ce atoms) and leaves a trans Te atom, bonded to Ce, available
to bind a third Ag atom. The latter acts to link the layers into a three-dimensional
structure (Figure 5.2C). It is interesting that if one removes the Ce atoms, the
remaining [Ag3Te4] substructure is still contiguous and three-dimensional. In this
sense, the Ce atoms occupy positions in an open silver telluride framework The
tunnels in the structure have an oval-shaped cross section with dimensions of
10.63A (Tel-Tel) x 5.63A (Te2-Te2) x 8.08A (Te4-Te4), see Figures 5.2D and
5.3. If the Van der Waals diameters are considered, the tunnels have an accessible

opening of 7.9A x 2.9A x 5.3A. These dimensions are large enough to suggest

244

that the K+ cations may be accessible via topotactic ion-exchange. The fractional
atomic coordinates, isotropic and anisotropic temperature factors, bond distances,
and bond angles for K2Ag3CeTe4 are given in Tables 5.4-5.6.

Structure Description of szCu3CeTe5 (II) - Rb2Cu3CeTe5 consists of
i[Cu3CeTe5]2' layers separated by Rb+ cations, see Figure 5.4. The Ce atom is
seven coordinate, exhibiting a distorted pentagonal bipyramidal geometry in which

one 112-(Te?) unit20 and three Te2' anions comprise the pentagon and two Tez'

anions occupy the axial positions, see Figure 5.5. The pentagonal bipyramids share

monotelluride ions, forming -1- [CeTes]5‘ chains parallel to the b—axis.
(I)

Conceptually, these one-dimensional chains derive from the ZrSe; structure type.
By replacing one (Q22) unit in the ZrSe3 fiamework with a QZ' unit, the
coordination environment of the metal changes from bicapped trigonal prismatic
to pentagonal bipyramidal. This change in coordination is accompanied by a

conversion from two-dimensional layers to one-dimensional chains. Within the

1
; [CeTe5]5' chains exist empty distorted tetrahedral pockets of Te atoms which are

large enough to accommodate Cu atoms. Each Cu atom is bonded at two points to
the axial positions of two neighboring pentagonal bipyramids, and at the

remaining sites to the closest edge between these axial positions. The chains, once

extended to include the Cu atom, can be written as :10-[Cu2CeTe5]3'. Finally, the

layers are formed when the second type of Cu atom "stitches" these chains

245

 

together in the a-direction by coordinating to neighboring chains in a distorted
tetrahedral arrangement. A view perpendicular to the layers is given in Figure
5.6A. It is interesting to note that if one removes the Ce atoms from the structure,

the remaining [Cu3Te3] substructure remains contiguous. In this sense, the Ce

atoms are situated on both sides of a two—dimensional [CuTe]" substrate. In fact,
this copper telluride fi'amework, albeit distorted, bears a close resemblance to the
layers of NaCuTe”, see Figure 5.6B. The fractional atomic coordinates, isotropic
and anisotropic temperature factors, bond distances, and bond angles for

szCu3CeTe5 are given in Tables 5.7-5.9.

246

 

 

 

 

Figure 5.1 ORTEP representation of the structure of KzAg3CCTC4 viewed down
the b-axis (90% probability ellipsoids). Ellipses with octant shading represent Ce,

crossed ellipses represent Ag, and open ellipses represent K and Te.

247

 

Figure 5.2 (A) Layers of KzCuZCeS4. (B) Corrugated [AngeTe4]3’ layers in
KzAgdCeTe4. (C) Inclusion of the third Ag atoms, between the [AngeTe4j3'
layers, links them together into a three-dimensional structure. The linking Ag
at01118 are highlighted by the shaded circle. (D) Tunnel window projection.

248

 

open channels in KZA CeTe4 with

Table 5.4 Fractional Atomic Coordinates and Equivalent Isotropic Displacement

Parameters (UN) for K2Ag3CeTe4 with Estimated Standard Deviations in

 

 

Parentheses.
atom x y z U 6,1“, A2
Ce(l) 0.3575(1) 0.2500 0.7714(1) 0.016(1)
Te(l) 0.2262(1) - 0.2500 0.7774(1) 0.017(1)
Te(2) 0.3608(1) 0.2500 0.5598(1) 0.022(1)
Te(3) 0.3769(1) 0.2500 0.9847(1) 0.020(1)
Te(4) 0.4835(1) - 0.2500 0.7506(1) 0.017(1)
Ag(1) 0.2779(1) - 0.2500 0.5978(1) 0.050(1)
Ag(2) 0.2726(1) - 0.2500 0.9525(1) 0.045(1)
Ag(3) 0.4679(1) - 0.2500 0.9301(1) 0.038(1)
K(l) 0.1 162(2) 0.2500 0.6558(2) 0.024(1)
K(2) 0.4288(2) - 0.2500 0.3991(2) 0.037(1)

 

aUeq is defined as one-third of the trace of the orthogonalized Uij tensor.

250

Table 5.5

Standard Deviations in Parentheses

Anisotropic Displacement Parameters (A) for KzAg3CeTe4 with

 

 

 

atom U11 U22 U33 U12 U13 U23
Ce(l) 0.0018(1) 0.0013(1) 0.018(1) 0.0000(0) 0.0000(1) 0.0000(0)
Te(l) 0.0017(1) 0.0015(1) 0.019(1) 0.0000(0) 0.0001(1) 0.0000(0)
Te(2) 0.0031(1) 0.0019(1) 0.016(1) 0.0000(0) 0.0001(1) 0.0000(0)
Te(3) 0.0027(1) 0.0017(1) 0.017(1) 0.0000(0) 0.0002(1) 0.0000(0)
Te(4) 0.0017(1) 0.0014(1) 0.021(1) 0.0000(0) 0.0002(1) 0.0000(0)
Ag(1) 0.0064(1) 0.0025(1) 0.060(1) 0.0000(0) 0.0027(1) 0.0000(0)
Ag(2) 0.0040(1) 0.0066(1) 0.028(1) 0.0000(0) 0.0001(1) 0.0000(0)
Ag(3) 0.0044(1) 0.0043(1) 0.028(1) 0.0000(0) 0.0007(1) 0.0000(0)
K(l) 0.0031(2) 0.0020(2) 0.022(1) 0.0000(0) 0.0001(1) 0.0000(0)
K(2) 0.0052(2) 0.003 1(2) 0.026(1) 0.0000(0) 0.0010(2) 0.0000(0)
The anisotropic displacement factor exponent takes the form: -2112 [(ha')2Un + + 2hka*b* U12]

251

Table 5.6 Selected Distances (A) and Bond Angles (deg) for K2Ag3CeTe4 with
Standard Deviations in Parentheses.

Bond Distances
Ce — Tel 3.2383(8) Ag2 — Tel 2.813(2) K1 — Te4 3.560(2)

Ce — Te2 3.261(1) Ag2 — Te2 2.828(2) K2 — Tel 3.888(3) '
Ce - Te3 3.303(1) Ag3 — Te4 2.788(1) K2 — Te2 3.536(3)
Ce — Te4 3.1898(9) Ag3 — Te3 2.922(1) K2 — Te4 3.677(3)
Agl — Tel 2.906(2) K1 — Tel 3.531(2)

Agl — Te2 2.785(1) K1 — Te3 3.513(2)

 

 

Bond Angles

Tel — Ce — Te2 9232(3) Te2 — Agl - Te2 112.81(6)
Tel - Ce — Te3 9242(2) Te2 - Agl — Te3 108.26(4)
Te2 — Ce — Te3 173.21(4) Tel — Ag2 — Te3 109.34(4)
Te4 — Ce — Tel 175.68(2) Tel — Ag2 — Te2 109.30(5)
Te4 - Ce — Te2 8354(3) Te2 — Ag2 — Te3 113.06(3)
Te4 - Ce - Te3 91 .81(3) Te3 — Ag2 — Te3 102.51(5)
Tel — Agl — Te3 105.37(5) Te3 — Ag3 — Te3 110.70(3)
Te2 — Agl — Tel 110.90(4) Te4 — Ag3 — Te3 109.77(4)

 

252

 

Figure 5.4 ORTEP representation of the structure of szCu3CeTe5 as seen down
the b.axis (90% ellipsoids). The ellipses with octant shading represent Ce and Rh,

the crossed ellipses represent Cu and the open ellipses represent Te.

253

o ' I '
‘0 0‘. ZI'SC3

('3 ."5 . :
. . .4. .4. .4. I,,..,,,.

 

Figure 5.5 Schematic comparison of the two-dimensional layers of ZrSe3, the

one-dimensional i[CeTe5]5' chains and the i[Cu2CeTe5]3' chains in

Rb2Cu3CeTe5. The dotted line highlights the pentagonal bipyramidal coordination

around Ce.

254

Cu “stitch”

 

 

Ease -mfiohoOwaOH
)
A b

L

CuTe—network

illustrating

Figure 5-6 (A) View perpendicular to the layers of RbZCu3CeTe5,

3' chains to form two-

[CU2CeTe5]

l
(1)

how the second Cu atom stitches together the

dimensional layers. The ditelluride groups above and below the anionic layers are

omitted for clarity. (B) The distorted [CuTe]', PbO-like layer in szCu3CeTe5.

255

Table 5.7 Fractional Atomic Coordinates and Equivalent Isotropic Displacement

Parameters (Ueq) for Rb2Cu3CeTe5 with Estimated Standard Deviations in

 

 

Parentheses.
atom x y z cha, A2
Ce(l) 0.2982(1) 0.0000(0) 0.1854(1) 0.009(1)
Te( 1) 0.3626(1) 0.2221(2) 0.4314(1) 0.016(1)
Te(2) 0.2777(1) 0.5000(0) 0.1052(1) 0.010(1)
Te(3) 0.1161(1) 0.0000(0) 0.1464(1) 0.011(1)
Te(4) 0.4604(1) 0.0000(0) 0.1453(1) 0.01 1(1)
Cu(l) 0.1421(1) 0.2521(4) 0.0009(2) 0.028(1)
Cu(2) 0.0000(0) 0.2492(4) 0.0000(0) 0.020(1)
Rb(l) 0.1732(1) 0.5000(0) 0.3192(2) 0.016(1)
Rb(2) 0.4962(1) 0.5000(0) 0.3148(2) 0.016(1)

 

“Ueq is defined as one-third of the trace of the orthogonalized Uij tensor.

256

Table 5.8

Standard Deviations in Parentheses.

Anisotropic Displacement Parameters (A) for szCu3CeTe5 with

 

 

 

atom U11 U22 U33 U12 U13 U23

Ce(l) 0.0009(1) 0.0008(1) 0.012(1) 0.0000(0) 0.0006(1) 0.0000(0)
Te(l) 0.0022(1) 0.0011(1) 0.014(1) 0.0003(1) 0.0006(1) 0.0001(1)
Te(2) 0.0011(1) 0.0008(1) 0.014(1) 0.0000(0) 0.0007(1) 0.0000(0)
Te(3) 0.0009(1) 0.001 1(1) 0.015(1) 0.0000(0) 0.0008(1) 0.0000(0)
Te(4) 0.0008(1) 0.0011(1) 0.016(1) 0.0000(0) 0.0008(1) 0.0000(0)
Cu(l) 0.0016(1) 0.0041(1) 0.031(1) 0.0008(0) 0.0012(1) 0.0022(0)
Cu(2) 0.0014(1) 0.0025(2) 0.021(2) 0.0000(0) 0.0011(1) 0.0000(0)
Rb(l) 0.0013(1) 0.0018(1) 0.020(1) 0.0000(0) 0.0009(1) 0.0000(0)
Rb(2) 0.0015(1) 0.0016(1) 0.019(1) 0.0000(0) 0.0010(1) 0.0000(0)
The anisotropic displacement factor exponent takes the form: -21t2 [(ha')2Un + + 2hka*b* U12]

257

Table 5.9 Selected Distances (A) and Bond Angles (deg) for szCu3CeTe5
with Standard Deviations in Parentheses.

Bond Distances
Ce—Tel 3.161(1) Cu2—Te3 2.721(2) Rbl—Te3 3.710(1)

Ce — Te2 3.2538(5) Cu2 — Te4 2.593(2) R61 — Te4 3.720(2)
Ce —- Te3 3.246(2) Cul - Cel 3.332(2) R62 — Tel 3.694(2)
Ce - Te4 3.253(2) Cul — Cu2 2.650(2) R62 — Te2 3.854(2)
Cul — T62 2.820(2) Tel — Tel 2.771(2) Rb2 — Te3 3.681(2)
Cul — Te3 2.591(2) R61 — Tel 3.707(2) R62 — Te4 3.683(1)
Cul - Te4 2.593(2) R61 - Te2 3.880(2)

 

 

 

Bond Angles

Tel — Ce — Tel 5199(4) Te2 — Cul — Te2 8870(6)
Tel — Ce — Te2 8054(3) Te3 — Cul - Te2 105.68(8)
Tel — Ce — Te3 9733(4) Te3 — Cul — Te4 125.03(9)
Tel — Ce — Te4 97 .00(4) Te3 — Cu2 — Te3 110.3(1)
Te2 — Ce — Te2 146.92(6) Te3 — Cu2 — Te4 103.21(4)
Te3 — Ce — Te2 87 .85(3) Te4 — Cul — Te2 106.62(8)
Te3 — Ce — Te4 164.05(5) Te4 — Cu2 — Te4 109.8(1)

Te4 — Ce — T62 8763(3)

 

258

Ion Exchange Pr0perties of K 2Ag3Ce Te, (1) - Solid-state ion-exchange
reactions22 were indeed performed in which the material was pressed with a fifty

fold excess of A1 (A = Li, Na, N114) and heated at 100°C for 5 days, see Scheme 1.

Scheme 1

K2Ag3CeTe4 + 50x AI 9.1“, M; 1% AXK2-xAg3CeTe4

The products were isolated by washing away the iodide matrix with methanol.
The resulting materials appeared to be isostructural as judged by powder x-ray
diffraction (Figure 5.7). Elemental analysis by EDS on the polycrystalline
material also appeared to support the premise that the potassium had exchanged
out of the channels in the framework, giving average compositions of
Li1.48K0.5243483C€T¢4s Na1_55K0,45Ag3CeTe4, 311d (NH4)1.35K0.65A83C9TC4- Since
EDS is a semiquantitative method, the polycrystalline materials were further
characterized by ICP to obtain accurate values for the formula. The ICP results for
the Na] reacted material were in agreement with those obtained by EDS giving a
formula of NaLzéKmsAgaCeTm. Considering that 'm typical ion-exchange
reactions, multiple cycles are required for complete exchange, where the material
has to be isolated and re-reacted with fiesh reagents several times, the observed
degree of ion-exchange in the first cycle for is remarkably high. Complete
exchange is expected in subsequent cycles. The results obtained by ICP for the UT

and N11,] reacted materials, however, did not agree with those results obtained by

259

EDS and seemed to indicate that the material exchanged to a much lesser degree.
The difference between the two methods is that EDS is surface technique while
ICP is a bulk technique. One postulation might be that the exchange for these
materials occurred on the surface only and that this exchange was followed by
decomposition. By EDS, only the surface of the material is probed which would
give rise to the above formulas. If the decomposition product were amorphous, it
would not show up in the powder x-ray diffiaction pattern. The isomorphous x-
ray powder diffraction pattern, however, would come from the amount of

unexchanged material within the core of each particle.

260

 

 

 

 

 

1
(A);
s? 1
g (3)1
1
1.
61 3
g; C1
a n ()‘1
$3 1
5'.
1 1
I 1
:' (”)1
‘10-“L20UIJ304L 40 50 +‘J6—0

20 angle (degrees)

Figure 5.7 Powder XRD patterns of (A) pristine K2Ag3CeTe4 before ion-
exchange, (B) Lil ‘1' K2Ag3CeTe4, (C) NH] '1' KzAg3CeTe4, and (D) NI‘LI +
K2Ag3CeTe4

261

Magnetic Susceptibility and Infrared Spectroscopy - The magnetic
susceptibility of K2Ag3CeTe4 was measured over the range 5-300K at 6000G. A
plot of 1/xM vs T shows that the material follows Curie-Weiss Law with only
slight deviation from linearity beginning below 50K, see Figure 5.8A Such
deviation has been reported for several Ce3+ compounds and has been attributed to
crystal field splitting of the cation’s ZFSQ ground state.23 At temperatures above

100K, a 1.1eff of 2.19 113 has been calculated, which is in accordance with the usual
range for Ce3+ compounds (2.3-2.5 118). The presence of Ce3+ is confirmed by

infra-red spectroscopy which shows five peaks at 3252 cm.1 (0.40 eV), 1648 cm'1
(0.20 eV), 1465 om" (0.18 eV), 1374 cm1 (0.17 eV), and 872 cm" (0.11 eV), see

Figure 5.88. These absorptions are electronic in origin and can be attributed to a f-

f and/or f-d transition within the f' configuration of Ce3+. From this data, it is
evident that the material is valence precise and the formal oxidation states can be
formalized as (Kl+)2(AgH)3(Ce3+)(TCZ')4o

The magnetic susceptibility of szCu3CeTe5 was measured over the
range 5-300K at 6000G, and a plot of l/xM vs T shows that the material exhibits
nearly Curie-Weiss behavior with only slight deviation from linearity beginning
below 50K, see Figure 5.9A. At temperatures above 150K, a pen. of 2.64113 has
been calculated, which is in accordance with the usual range for Ce3+ compounds

(2.3-2.51113). The diffuse reflectance spectra was measured in the Mid-IR reg'on

for Rb2CU3CCT65 and is shown in Figure 5.9B. One broad, well-defined, peak is

262

present at 3355cm'l (0.42 eV), which could be attributed to a f-f and/or f-d
transition within the 1" configuration of Ce”. Another explanation may be that that
water is being absorbed onto the crystals and this peak is coming from the O-H
bending mode of water.

From this we can conclude that Rb2Cu3CeTe5 is a valence precise

compound, and thus should expect semiconducting behavior. The formal

oxidation states can be formalized as (Rbl+)2(Cu153(Ce3+)(Te2')3(Te22').

263

 

l/x

 

 

700

_ (A)

r11 I I
O

llllUlllll

500'- ,'

l LlLl I

400

300

III—IIII‘TIIII
O
I‘LL]

200:;- e

Ill.-

100 :- 0. -

 

 

O llIllnnnnlLlllllllllnnIllnnjn

0 50 100 150 200 250 300
Temperature (K)

 

 

2.0 I I I I I I 1 I I I—' I 1 I I I I r11 I 1 'TI I I

(B)

  

1.8 '
1.6
1.4
1.2

% Reflectance

1.0
0.8
0.6

JljllllllLlJl—IIJJLLLLLLLLJJ

800 1600 2400 3200 4000 4800 5600
Wavenumber (cm‘l)

Figure 5.8 (A) Inverse molar magnetic susceptibility (l/xM) plotted against
temperature (2-300K) for KzAg3CeTe4. (B) Diffuse reflectance optical spectra of
KzAg3CeTe4 (in the Mid-1R region).

264

 

 

500.00

400.00 b o

2 300.00 " '

l/x

200.00 1. 0

1111.00 ” 0

 

l l l l 1 LJ 1 l l j

 

 

0m nnnnlnnnnanLnlnn
0 50 100 150 200 250 300

Temperature (K)

 

2.0

II'IIT'IIIIIIITr—FITTTjII

«113....

1.5

% Reflectance

1.0 ‘

1J4 14l+l+l41_14 LMJLLI l l lel
0.5

5600 4800 4000 3200 24001600 800
-l
Wavenumber (cm )

Figure 5.9 (A) Inverse molar magnetic susceptibility (l/xM) plotted against
temperature (2-300K) for szCu3CeTe5. (B) Diffuse reflectance optical spectra of
szCu3CeTe5 (in the Mid-IR region).

265

Charge Transport Properties - Electrical conductivity data as a function
of temperature for both a single crystal and a room temperature pressed pellet of
KzAg.,CeTe4 show that this material is a narrow gap semiconductor with room
temperature values ranging from 0.1 S/cm for the single crystal to 0.01 S/cm for
the pressed pellet, see Figure 5.10A. A bandgap of 0.36 eV was obtained by
fitting the single crystal conductivity data to the semiconductor equation.24 The
thermopower data for two single crystals of KzAg3CeTe4 show very large Seebeck
coefficients ranging from 500 to 700 uV/K, see Figure 5.10B The positive sign
and decreasing Seebeck coefiicient with falling temperature are also consistent
with a p-type narrow gap semiconductor.

The electrical conductivity of szCu3CeTe5 as a function of temperature
measured on single crystals suggests that the material is also a narrow gap
semiconductor with a room temperature value of 0.05 S/cm, see Figure 5.11A. A
plot of logo vs l/T is nonlinear over the entire temperature range of 8-300 K,
suggesting the conduction mechanism varies in different temperature regions,
possibly due to different type of mid gap states. Thermoelectric power data as a
function of temperature show a large Seebeck coefficient at room temperature of
+275 uV/K, see Figure 5.11B. The increasing Seebeck coefficient with
decreasing temperature and the positive sign are consistent with a p—type

semiconductor.

266

10'1

10'2

Log 0’ (S/cm)
8
L

SUN/K)

Electrical Conductivity

 

 

 

 

 

 

 

I I I Q I I l j j l 1 l4 1 l #1 l l l l
100 150 200 250 300
Temperature (K)
Thermopower
Y I 1 i I jiT I I T I I T

crystal _-
‘ v W i
.‘:
omquw ‘m. A -2. 2.. CTYStal %
“W " “ “M 3
.1

1 14 141 1_1 Li #1414 1_J__l_1_1__1 _1_L_1_L

 

 

180 200 220 240 260 280 300
Temperature (K)

Figure 5.10 (A) Variable temperature, four probe electrical conductivity data for

a single crystal and a pressed pellet of KzAg3CCT84. (B) Variable temperature

thermopower data for single crystals of KzAggCeTelt.

267

-1 Electrical Conductivity

10
,2 (A)
10

 

10'3

Log 0' (S/cm)
3“

 

 

 

10*5
10'7
-3 9
10 o
10'9jILLIIJIAIILLLLIIIAIIllllllll
0 50 100 150 200 250 300
Temperature (K)
Thermo wer
350* EX) .
(B) 5
. mwf‘ . 0.;.»“9-1
300 e . VOW, -

SUN/K)

- 1
250 - é

 

 

 

200.- . I+Lil4k14l lJLlJll—l IIJJJLLIL+

160 180 200 220 240 260 280 300
Temperature (K)

Figure 5.11 (A) Variable temperature, four probe electrical conductivity data
for a single crystal of szCu3CeTes. (B) Variable temperature thermopower data
for a single crystal of szCu3CeTe5.

268

D. Conclusions

In summary, two new quaternary tellurides, K2Ag3CeTe4 and RbZCu3CeTe5,
have been discovered by combining either copper or silver with cerium in alkali
metal/polytelluride fluxes. Both compounds are “coinage metal rich” in their
formulas and possess new structure types. While K2Ag3CeTe4 is very unique in
that it can undergo topotactic ion-exchange with sodium due to the large open
cavities in its structure, szCu3CeTe5 is interesting in that it is structurally related
to both CuTe and CeTe3. Both compounds are valence precise, and therefore
behave as semiconductors. However, optical bandgaps were not able to be
determined for either of these compounds due to the presence of f-f and/or f-d
transitions that mask these regions in the diffuse reflectance IR spectra.

It is interesting to note that the formulas of these compounds are very similar,
differing only by one Te. From this, one might question whether or not the
reverse compounds, KzAg3C6T65 and/or szCu3CeTe4, exist. There was no
evidence for the formation of either of these compounds, even though both formed
under relatively similar synthetic conditions. This suggests that the structure types
observed here are particularly unique and quite possibly only stabilized under

these specific combination of elements.

269

 

 

References

l

10

13

(a) Kanatzidis, M.G.; Sutorik, A.C. Prog. Inorg. Chem. 1995, 43, 151 and
references therein. (b) Pell, M.A.; Ibers, J .A. Chem. Ber./ Recueil 1997,
130, 1.

Norling, B.K.; Steinfink, H. Inorg. Chem. 1966, 5, 1488.

Krbnert, V.W.; Plieth, K. Z. Anorg. Allg. Chem. 1965, 336, 207.

Sutorik, A.C., Kanatzidis, M.G. Chem. Mater. 1997, 9, 387.

(a) Cody, J.A.; Ibers, J.A. Inorg. Chem. 1996, 16, 3273. (b) Wu, P.; Pell,
M.A.; Ibers, J.A. J. Alloys and Compd. 1997, 255, 106. (c) Choi, K.-S.;
Patschke, R; Billinge, S.J.L.; Waner, M.J.; Dantus, M.; Kanatzidis, M.G. J
Am. Chem. Soc. 1998, 120, 10706.

Patschke, R.; Heising, J .; Schindler, J .; Kannewurf, C.R., Kanatzidis, M.G.
J. Solid State Chem. 1998, 135, 111.

Sutorik, A.C.; Albritton-Thomas, J .; Kannewurf, C.R.; Kanatzidis, M.G. J.
Am. Chem. Soc. 1994, 116, 7706.

Sutorik, A.C.; Albritton-Thomas, J.; Hogan, T.; Kannewurf, C.R.,
Kanatzidis, M.G. Chem Mater. 1996, 8, 751.

Cody, J.A.; Ibers, J.A. Inorg. Chem. 1995, 34, 3165.

(a) Wu. P.; Ibers, J.A. J. Solid State Chem. 1993, 110, 156. (b) Christuk,
A.E.; Wu, P.; Ibers, J.A. J. Solid State Chem. 1994, 110, 330. (c) Wu. P.;
Ibers, J .A. J. Solid State Chem. 1994, 110, 337.
Huang, F. Q.; Choe, W.; Lee, S; Chu, J.S. Chem. Mater. 1998, 10, 1320.
Bensch, W.; Dilrichen, P. Chem. Ber. 1996, 129, 1489.

Patschke, R.; Brazis, P.; Kannewurf, C.R.; Kanatzidis, M.G. Inorg. Chem.
1998, 37, 6562.

270

 

14

15

16

17

18

19

20

21

22

23

24

Patschke, R.; Brazis, R.; Kannewurf, C.R.; Kanatzidis, M.G. J Mater.
Chem. 1998, 8, 2587.

SMART: Siemens Analytical Xray Systems, Inc., Madison, WI, 1994.

SAINT: Version 4.0, Siemens Analytical Xray Systems, Inc., Madison WI,
1994-1996.

SADABS: Sheldrick, GM. University of G6ttingen, Germany, to be
published.

Sheldrick, GM. SHELXTL, Version 5; Siemens Analytical Xray Systems,
Inc.; Madison, WI, 1994.

Stowe, H.D.; Braselton, W.B.; Kaneene, J .B.; Slanker, Am. J. Vet. Res
1985, 46, 561.

The Te-Te stretch exhibits a Ramarl shifi at ~160 cm'l.
Savelsberg, G; Schafer, H. Z. Natwforsch, B. 1978, 33b, 370-373.

Chondroudis, K.; Kanatzidis, M.G. J Am. Chem. Soc. 1997, 119, 2574. (c)
Chondroudis, K.; Kanatzidis, M.G. J Solid State Chem. 1998, 136, 328.

(d) Hanko, J .A.; Kanatzidis, M.G. Angew. Chem. Intl. Ed Engl. 1998, 37,
342.

Greenwood, N.N; Eamshaw, A. Chemistry of the Elements; Pergamon
Press: New York, 1984; pl443.

(a) Smith, R.A.; Semiconductors, 2nd ed.; Cambridge University Press:
Cambridge, New York, 1978; p19. (b) Wold, A.; Dwight, K. Solid State
Chemistry: Synthesis, Structure, and Properties of Selected Oxides and
Sulfides; Chapman & Hall: New York, 1993, p 35.

271

-"fl'l'f ’

Chapter 6
Cu,UTe3 (x = 0.25 and 0.33): Stabilization of UTe, in the

ZrSe, Structure Type via Copper Insertion

272

A. Introduction

Recently, a survey of the structural chemistry of both ternary and
quaternary uranium (and thorium) chalcogenides was presented.1 Among the most
notable in this class include CsUTe6,2 ngHfsUTe3o_6,2 AMUQ3 (A = alkali or
alkaline earth metal, M=3d metal, Q = 3, Se, Te),2b’3 CsTiUTe5,2b T10_56UTe3,4
KZUP3Seo,5 and Rb4U4P4Se466. Even more recently, we have described the three
novel quaternary uranium chalcogenides, K6Cu12U2S15,7 KUzsbSCg,8 and
RbUz Sng 8_ Here, we report on our investigations into the copper uranium
telluride system which afforded the interesting new compound, CuxUTe3 (x = 0.25
and 0.33). Only two other ternary copper uranium chalcogenide phases have been
reported (i.e.: Cu2U3Q79 and Cu2U6Q13 (Q=S, Se)'°) which were found in the
sulfide and selenide systems. Although formulated differently, CuxUTe3 (x = 0.25
and 0.33) is isostructural to the previously reported phase, CuThzTe6” and adopts
the layered ZrSe3-structure type. Structurally, these ternary compounds derive
from the parent binary layered phases by inserting copper atoms between the
layers. The occupancy of the copper site ranges from 0.25 to 0.50. In this sense,
these compounds can be compared to the intercalation compounds, LierQ3 (0 < x
S 3).”’13 The insertion of a metal atom between these ZrSeg-type layers is, in fact,
not unusual. In addition to the LierSe3 (0 < x _<. 3) phases, a series of compounds
with the formula ATh2Q5 (A = K. Rb. and Cs; Q = $9 and Te)14 has been reported
where an alkali metal cation has been inserted between the ZrSe3-type layers, now

ThQ3 (Q = Se and Te). Here, we report on the structure and physicochemical

273

properties of CuxUTe3 (x = 0.25 and 0.33) and discuss it with respect to the parent
binary phase, a—UTe3.
B. Experimental Section

1. Reagents. The following reagents were used as obtained: (i) copper
powder, 99.9% pure, Fisher Scientific Co., Fairlawn, N.J. (ii) uranium powder,
99.7% pure, 60 mesh, Cerac, Milwaukee, WI; (iii) tellurium shots, 99.9% pure,
Noranda Advanced Materials, Saint-Laurent, Quebec, Canada.

2. Synthesis — All manipulations were carried out under a dry nitrogen
atmosphere in a Vacuum Atmospheres Dri-Lab glovebox.

CuxUTe3 (x = 0.25 and 0.33). Amounts of 0.076 g (3.0 mmol) of Cu,

0.095 g (1.0 mmol) of U, and 0.204 g (4.0 mmol) of Te were weighed into a vial
in an Nz-filled glovebox. The starting materials were mixed thoroughly and
loaded into a carbon-coated silica ampoule. The ampoule was then evacuated to <
1 x 10‘4 mbar and flame-sealed. In a computer-controlled furnace, the reaction
was heated to 800°C over 36 hours, held at that temperature for 6 days, cooled to
400°C at 4°C/h, further cooled to 100°C at 6°C/h and quenched to 50°C. The
ampoule was opened in air to reveal the product, which consisted of purple cubes
(25%), black powder (25%) and silver needles and plates (50%). All entities of
the product are air- and water-stable. The purple cubes and black powder were
identified by semiquantitative energy dispersive spectroscopy (EDS) to be CuzTe
and UTe3, respectively. The silver needles and plates gave the same average

composition of Cuo,25Ul,oT€/27-

274

3. Physical Measurements - The instrumentation and experimental setup
for the following measurements are the same as described in Chapter 2, Section 3:
Semiquantitative Energy Dispersive Spectrosc0py (EDS), Powder X-ray
Diffraction, Transmission Electron Microscopy, and Charge Transport
Measurements.

X-ray Crystallography — For reasons outlined in the results and discussion
section, several crystals were examined crystallographically. Crystal #1: A single
crystal with dimensions of 0.02 x 0.05 x 0.10 mm was mounted on the tip of a
glass fiber. Intensity data were collected at room temperature on a Rigaku AF C6S
four-circle automated diffi‘actometer equipped with a graphite-crystal
monochromator. The unit cell parameters were determined from a least-squares
refinement using a setting angles of 20 carefirlly centered reflections in the 8° 3 20
3 30° range. The data were collected with an 01-20 scan technique over one-
quarter of the sphere of reciprocal space, up to 60° in 20. Crystal stability was
monitored with three standard reflections whose intensities were checked every
150 reflections. No significant decay was detected during the data collection
period. An empirical absorption correction based on w-scans was applied to all
data during initial stages of refinement. The structure was solved by direct
methods using SHELXTL'S package of crystallographic programs. Crystals #2
and #3: Single crystals with dimensions of 0.03 x 0.05 x 0.08 mm for crystal #2

and 0.03 x 0.04 x 0.10 mm for crystal #3 were mounted on the tip of a glass fiber.

275

“' “ _ ‘— mm ~cu—.'-‘"

Intensity data were collected at 173.1K for Crystal #2 and room temperature for
Crystal #3 on a Siemens SMART Platform CCD diffractometer using graphite
monochromatized Mo K01 radiation. The data were collected over a full sphere of
reciprocal space for both crystals, up to 56° in 20. The individual frames were
measured with an 00 rotation of 03° and an acquisition time of 603cc for crystal #2
and 30sec for crystal #3. The SMART16 software was used for the data
acquisitions and SAINT ‘7 for the data extractions and reductions. The absorption
corrections were performed using SADABS.18 The structures were solved by
direct methods using the SHELXTL'S package of crystallographic programs. The
complete data collection parameters and details of the structure solutions and

refinements for all three crystals are given in Table 6.1.

276

 

Table 6.1. Crystallographic Data for CUXUT€3 (x = 0.25 and 0.33)

 

 

Crystal #1 Crystal #2 Crystal #3
Chemical Formula Cu(mUTe, Cu(mUTe, CUoggUTeg
crystal habit, color needle, black needle, black needle, black
Diffractometcr Rigaku AFC6S Siemens SMART Siemens SMART
Platform CCD Platform CCD
Radiation Mo-Ka (0.71073A) Mo-Ka (0.71073A) Mo-Ko. (0.71073A)
Crystal Size, mm3 0.02 x 0.05 x 0.10 0.03 x 0.05 x 0.08 0.03 x 0.04 x 0.10
Temperature, K 293 173 293
Crystal System Monoclinic Monoclinic Monoclinic
Space Group P2,/m (#11) P2./m (#11) P21/m (#11)
a, A 6.0944(11) 6.0901(12) 6.0838(12)
b, A 4.2158(11) 4.2083(8) 4.2140(8)
c, A 10.3668(9) 10.335(2) 10.361(2)
13, deg 98.87400) 9895(3) 9883(3)
v, A 263. 16(9) 261 66(9) 262.47(9)
Z 2 2 2
u, mm'l 47.936 48.529 48.063
indexranges 05h59 05h57 -85h57
05k56 -55k55 ~55k55
-1551515 -l351513 4351513
20..., deg 60 56 56
total data 1065 692 2521
unique data 1064 691 692
R(int) N/A N/A 0.0439
no. parameters 32 32 32
final Rl/wR2', % 626/2008 5.04/1 1.80 4.13/10.42
Goof 1.139 1.047 1.190

 

'Rl=2(1Fol - IF.l)/>:|F.T wR2= {EMFoz-Fffl/Z[WtFozflim

All structures were solved and refined using the SHELXTL—S package of crystallographic programs.ls
SI-IELXTL refines on F2. An empirical absorption correction was performed during the initial stages of
each refinement (based on ‘P-scans for Crystal #1 and SADABS18 for Crystal #2 and #3).

277

C. Results and Discussion

Structure Description - The observed crystal structure of CuxUTe3 (x = 0.25
and 0.33) viewed down the b-axis is shown in Figure 6.1. The three-dimensional
framework is built fi'om layers very similar to those found in ZrSe3, which are
linked together by copper atoms. In a-UTeg, which adopts the ZrSe3-structure
type, each U atom is coordinated to eight Te atoms in a bicapped trigonal
prismatic environment. These trigonal prisms stack in one-dimension to form
wedge-shaped columns by sharing triangular faces. Layers are then formed when
neighboring columns share both their capping and apex monotellurides, see Figure
6.2A. Within these layers, there are ditelluride units that orient with their Te-Te
bonds parallel to the a-axis. The Te-Te bond distances are 2.751(1)A within the
ditelluride units and 3.350(1) A between them. In CuxUTe3 (x = 0.25 and 0.33),
however, the Te-Te distances (Te2-Te3 = 3.098(2)A, Te3-Te2 = 2.987(2) A) are
almost equal, giving rise to infinite chains running along the [100] direction. The
copper atom is stabilized in a distorted tetrahedral geometry and sits on a mirror
plane, which generates a pair of crystallographically related sites. The distance
between these two sites is 2.556A. Although this Cu—Cu distance is reasonable,
the copper atoms probably do not sit on both sites at the same time due to the
partial occupancy on this site. The fractional atomic coordinates, isotropic and
anisotropic temperature factors, bond distances, and bond angles for Crystal #3 are

given in Tables 6.2-6.4.

278

‘3“ ‘L— -L_— szmvamuh

The cell parameters of CuxUTe3 (x = 0.25 and 0.33), compared to those of
a-UTe3, show only a slight expansion along the c-axis of 0.0548A and a slight
increase in the volume of 0.27A3. Consequently, there is essentially no shift in the
positions of the peaks in the x-ray powder diffraction pattern. As a result,
CuxUTe3 (x = 0.25 and 0.33) cannot be readily distinguished from the a—UTe3 by
casual comparison of the two patterns.

The structure of CuxUTe3 (x = 0.25 and 0.33) is similar to that of
Tlo_5(5UTe34 (Figure 6.3). It is therefore instructive to compare these two structures
and understand the differences they pose. Both compounds are built fiom layers
of o1-UTe3 with metal atoms inserted between them on partially occupied sites.
The difference, however, lies in both the way that the We. layers stack with
respect to one other and how the metal cations insert between these layers. In
CuxUTe3 (x = 0.25 and 0.33), the layers of UTe3 stack in such a way that
tetrahedral pockets are formed between the layers for the copper atoms to reside.
In T1056UTe3, the layers shift with respect to one another so that a larger, square
prismatic pocket is formed for the much larger thallium atom to reside. This shift
in the layers completely changes the symmetry of the compound. While CuxUTe3
(x = 0.25 and 0.33) remains isostructural to the or-UTe3 (monoclinic), T1056UTe3 is

orthorhombic.

279

3.0978(16)A 2.9875(16)A

Figure 6.1 ORTEP representation of the structure of CuxUTe3 (x = 0.25, 0.33)
as seen down the b-axis (80% ellipsoids). The ellipses with octant shading
represent U atoms. The crossed ellipses represent Cu atoms and the open ellipses

represent Te atoms.

280

Table 6.2.
Displacement Parameters (A2 x 103), and occupancies for Cuo,25UTe3 (Crystal #3)
with Estimated Standard Deviations in Parentheses.

Fractional Atomic Coordinates ( x 104) , Equivalent Isotropic

 

 

atom x y z Ueqa, A2 Occ.
U 0.7914(1) 1/4 0.1629(1) 0.0009(1) 1
Te(l) 0.2659(2) 1/4 0.0599(1) 0.0009(1) 1
Te(2) 0.4007(2) 1/4 0.6620(1) 0.0014(1) 1
Te(3) 0.9114(2) 1/4 0.6685(1) 0.0016(1) 1
Cu 0.0930(15) 1/4 0.4656(7) 0.0019(2) 0.25

 

"‘U,‘l is defined as one-third of the trace of the orthogonalized Uij tensor.

Table 6.3
#3) with Estimated Standard Deviations in Parentheses

Anisotropic Displacement Parameters (A2) for CumsUTegt (Crystal

 

 

U11 U22 U33 U12 U13 U23
U 0.0008(1) 0.0005(1) 0.0013(1) 0 0.0001(1) 0
Te(l) 0.0007(1) 0.0009(1) 0.0011(1) 0 0.0001(1) 0
Te(2) 0.0016(1) 0.0011(1) 0.0017(1) 0 0.0007(1) 0
Te(3) 0.0019(1) 0.0011(1) 0.0018(1) 0 0.00040) 0
Cu 0.0034(5) 0.0013(4) 0.0012(3) 0 0.0010(4) 0

 

fie anisotropic displacement factor exponent takes the form: -21t2 [hza2 U11 +
+ 2hka b U12]

281

Table 6.4 Selected Distances (A) and Bond Angles (deg) for Cuo,25UTe3 (Crystal
#3) with Standard Deviations in Parentheses

 

 

Bond Distances

U — Tel 3.1026(10) x 4 Cu — Te2 2.509(4) x 4

U - Te2 3.1272(13) x 4 Cu — Te3 2.540(9) x 2
U—Te3 3.1193(11)x4 Te2—Te3 3.1019(19)x2
Bond Angles

 

Tel — U — Tel 85.47(3) 7601(3) 76. 14(3) 141 .75(4)
Tel — U — Te2 151.72(3) 8798(3) 7558(3) 128.73(2)
Tel — U — Te3 l49.46(4) 8698(3) 130.24(3) 7345(3)
Te2 — U — Te3 57.10(3) 8466(4) 111.45(3)

Te2 - Cu — Te3 72.3(2) 112.6(2)

Te3 — Cu — Te3 119.4(2) 113.3(3)

 

282

ZrSe3 type (01-UTe3) NdTe3 type (B-UTe3)

 

 

l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.2 Extended stuctures of (A) a-UTe3 and (B) B-UTe3.

283

 

 

 

 

 

 

 

 

 

 

 

 

Figure 6.3 Extended structure of T10.56UTe3 as seen down the b-axis.

284

611- vs ,B-type UT e3. After the crystals of CuxUTeg, (x = 0.25 and 0.33) were
discovered in the reaction mixture, efforts were made to synthesize the compound
as a single phase through a rational synthesis. Since reactions of direct
combination led to a mixture of CuxUTe3 (x = 0.25 and 0.33) as well as both the
01- and B-type19 UTe3, we decided to prepare 01-UTe3 as a starting material for
further reaction with copper in a second step. The problem that we encountered
was that the a-UTe3 is less thermodynamically stable than the B-UTe3, making it
difficult to prepare pure. The structural difference between the 01- and B-UTe; lies
in the coordination environment of the uranium atoms. As previously described,
a—UTe3 (Figure 6.2A) consists of uranium atoms that are eight coordinate
bicapped trigonal prismatic with rows of ditelluride atoms above and below the
layers of uranium atoms. In B-UTe3, which adopts the NdTe3 structure type (see
Figure 6.2B), the uranium atoms expand their coordination sphere to include nine
Te atoms in a tricapped trigonal prismatic arrangement. B-UTe; is structurally
more dense and, as a result, the tellurium atoms above and below the plane of
uranium atoms are best described as a square Te net. The literature reports the

following synthesis to make or-UTe3.20

650°C 650°C
U + 3Te ____* grind ____.> a-UTe3 (eql)
1 week 1 week

285

Our attempts to reproduce this synthesis, however, resulted only in B-UTe3. An
added complication derived from the fact that there exist several other UxTey
binary compounds with similar compositions (i.e.: UTe._g7,2' UTe2,2°’22
UTe3_;...,2°’2"23 Ute...”24 UzTe3,25 UTe5,2°’26 U2Te5,2°’27 U3Te5,28 and U7Te1229). In
order to avoid these binary phases, a seriés of reactions were run with a UzTe ratio
of 1:2.5 and the products were monitored as a function of time over the course of 5
days while heating at 525°C. After one day, the product was determined by
powder X-ray difli’action to be pure a-UTe3. The powder patterns surprisingly
did not change up to 5 days. These results indicate that at a 1:2.5 ratio, a-UTe3
will consistently form as a pure product. In fact, it does not matter which
temperature is chosen for this reaction to occur. As long as the ratio is 1:2.5, the
mixture can be heated as high as 900°C for 7 days and or-UTe; will form as a pure
product. When the ratio is changed to 1:3, however, the results were quite
different. A second series of reactions were performed where U and Te were
mixed in a ratio of 1:3 and heated to 650°C from 1 day to 11 days, see Figure 6.4.
After two days, the product was a mixture of the 01- and B-UTe3. After 5-7 days,
the product was still a mixture but the peaks corresponding to the B-UTe3 grew in
intensity while those corresponding to the o1-UTe3 decreased in intensity. After 11
days, the product consisted of B-UTe3 only. Finally, a set of experiments was
conducted where the U to Te ratio was chosen to be 1:2.5, 1:3.0, 1:3.5, 1:4.0, and

1:4.5. The results are summarized in Table 6.5. From these experiments, it can

286

be concluded that a-UTe3 is less thermodynamically stable than B-UTe; at a U:Te
ratio of 1:3 and that the product formed depends more strongly on the amount of
tellurium added rather than the temperature or time chosen for the reaction to

occur.

287

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EA)
ZrSe3-type
UTe3
(B) .
o o
2 days
(C)
00 °
5 days
(D)
7 days
11 days

 

50 60
o ZrSe3-type UTe3 o NdTe3-type UTe3

Figure 6.4 X-ray powder diffraction patterns of (A) a-UTe3 and (B)-(E) the
products of 1U + 3Te heated to 650°C for 2days, 5days, 7days, and 11 days.

288

Table 6.5 Relative Stability of the UTe3 structure types as a function of amount

of tellurium added. The reaction was heated to 650° C for one week.

 

 

Reaction Product(s)

1U + 2.5Te o1 - UTe3
1U+3.0Te a-andB-UTe3
1U + 3.5Te o1 - and B - UTe3
1U + 4.0Te B - UTe3

1U + 4.5Te B - UTe3

 

The products were determined by powder x-ray diffi'action

289

 

Once the or-UTe3 was prepared pure, it was used as a starting material for

further reaction with copper.
xCu + 111-UTe3 ——'> CuxUTe3 (eq. 2)

Mixtures of Cu and 111—UTe3 in the ratio of 0.25, 0.33, 0.5, 0.75, 1.0, 1.25, and 1.5
to 1.0 were pressed into pellets and heated at 300° C for 2 days in a 13mm pyrex
ampoule that was flame sealed under vacuum. (Note: the Cu metal was first
activated by filtering it with copious amounts of dilute hydrochloric acid. If this
step is not taken, the oxide coating on the Cu metal prevents it from reacting with
the a—UT e 3). The idea was that under mild heating conditions and close physical
packing, the copper would be able to insert between the layers of UTe3 and
transform to the ternary CuxUTe3 (x = 0.25 and 0.33) phase. As discussed earlier,
there is no recognizable difference in the positions of the peaks in the x-ray
powder pattern of CuxUTe3 (x = 0.25 and 0.33) compared to that of 111-UTe3.
However, if the peaks corresponding the elemental copper, which are clearly
distinguishable, decrease in intensity or even disappear, this could be evidence for
the fact that the copper has inserted between the layers of a-UTe3 and transformed
to CUXUTC3 (x = 0.25 and 0.33). The powder patterns of the elemental Cu, the

reaction of 0.5Cu + 1.0 a—UTe3 before heating, and the product of 0.5Cu + 1.0

a—UTe3 after heating are shown in Figure 6.5. The peaks from the elemental

copper have noticeably disappeared in the product, suggesting that the Cu has

successfully been inserted between the layers. The EDS analysis gives an average

290

 

composition of CqugUTe3, confirming that there is indeed copper in the product.
In order to really determine whether or not the Cu has inserted between the layers,
however, more evidence is needed. We are currently in collaboration with
Professor Simon Billinge (Dept of Physics, Michigan State University) to study
these samples by PDF (pair distribution function). PDF analysis could potentially
be very helpfirl since this technique is capable of probing for and detecting Cu-Te

bond distances.

291

 

Intensity (arbitrary units)

"ken—P .m;n

 

fiUrIII'I'l

Copper Metal

 

 

 

IfT—l'j' I

'l

I'Irr' VI

.1

0.5 CI] '1' UTB3
before heating

 

 

l Ajjllj

 

 

 

 

d.

0.5Cu + UTe3
after heating

 

- “an--- ... 1 n n .-.n-th.A-H

30 40 50 60

Figure 6.5 Powder x-ray diffraction patterns of (A) elemental copper, (B) 0.5
Cu + 1.0 or—UTe3 before heating, and (C) 0.5 Cu + 1.0 a—UTe; after heating.

292

Superstructure - In order to balance the charges of CuxUTe3 (x = 0.25 and
0.33), one must understand how the insertion of copper has affected the UTe3
fi’ameworlc In order to accommodate the extra +0.25 and +0.33 charge, some
atoms in the framework must be reduced. Due to the close proximity of the
infinite Te chains in the structure to the copper atoms, it is most likely that these
Te atoms are acting as the electron acceptors. This means that some of the Tezz'
units in the parent UTe3 structure will be reductively cleaved as indeed is observed
crystallographically, see Figure 6.1 and Table 6.4. Therefore, a reasonable
formula would be (Cu+)x(UM)(Te2')(Te""))2 and it is possible that such a reduction
could cause a subtle superstructure to form whereby some of the Te-Te bonds in
the chain are broken. Perhaps a clue for the presence of a superstructure comes
from the anisotropic temperature factors, U11 and U33, for the Te atoms in the
chains (Te2 and Te3), which are larger than those of the uranium and Tel atoms,
see Table 6.3.

Transmission Electron Microscopy - To probe for such a superstructure, we
used electron diffraction. Since it was not yet clear as to how the x-value of
CUXUTC3 (x = 0.25 and 0.33) would affect any present modulation, the same
crystals used for the x-ray structure determination (Crystal #2 and #3) were
carefully removed from the glass fiber and prepared for study by transmission
electron microscopy (TEM). Both crystals showed evidence for a superstructure,
see Figures 6.6 and 6.7. In fact, two different superstructures were found. When

the amount of copper in the compound was 0.25, an incommensurate 6. 25am x

293

 

9..." .J-"“- o

 

1b,“), supercell resulted while a commensurate supercell of 6a,“), x 1b,"), was found
when the amount of copper was 0.33. These results tell us that the amount of
copper introduced between the layers directly affects how the Te chains in the

structure distort by dictating how many Te22' units are being reduced.

294

 

 

 

Figure 6.6 (A) Selected area electron diffi'action pattern of Cuo,25UTe3 with the
electron beam perpendicular to the layers ([001] direction) showing the
incommmsurate superlattice reflections along the a*-axis. (B) Densitometric
intensity scan along the a*-axis of the electron diffraction pattern (boxed area on
Photograph) showing the (h10) family of reflections. The three reflections from
the sublattice of CumsUTeg, are indexed. The four weak peaks are from the
suPerlattice with asuper = 6.25am,

295

Electron Diffraction of Cu0_25UTe3

 

 

Intensity (arbitrary units)

 

 

Reciprocal Angstroms

296

 

.. u ‘
A _‘ — _. ‘u ...1- A final”- __—. :2

Figure 6.7 (A) Selected area electron diffraction pattern of Cu0,33UTe3 with the
electron beam perpendicular to the layers ([001] direction) showing the
incommensurate superlattice reflections along the a*-axis. (B) Densitometric
intensity scan along the a*-axis of the electron diffraction pattern (boxed area on
Photograph) showing the (h10) family of reflections. The three reflections fi'om
the sublattice of Cuo_33UTe3 are indexed. The four weak peaks are from the
s“Perlattice with am,at = 6.0asub_

297

Electron Diffraction of Cuo.33 UTe3

 

 

Intensity (arbitrary units)

 

 

 

Reciprocal Angstroms

298

 

 

Charge Transport Praperties - Charge transport measurements were made
on bulk crystals of CuxUTe3 (x = 0.25 and 0.33) as well as on a polycrystalline
pressed pellet of the binary a-UTe3. The electrical conductivity and thermopower
data on CuxUTe3 (x = 0.25 and 0.33) are shown in Figure 6.8. The room
temperature conductivity reaches ~280 S/cm and decreases with decreasing
temperature, suggesting a semiconductor. At 250K, there is an anomalous dip in
the data. Interestingly, this dip also exists in the thermopower data at the same
temperature. While we are unsure of the cause of this anomaly, we are certain that
it is not due to a structural transition since single crystal x-ray data was collected
for this compound both above and below this temperature and the same
crystallographic structure was observed (see Table 6.1). The electrical
conductivity of a-UTe3 also indicates semiconducting behavior with a room
temperature value of 10 S/cm, ahnost 30 times less than CUXUTC3 (x = 0.25 and
0.33), see Figure 6.9. However, because these measurements were made on a
pressed pellet this drop in conductivity might be largely due to grain boundary
affects.

The thermopower data of CuxUTe3 (x = 0.25 and 0.33) suggests that the
material is a p-type semiconductor down to 40K. Below this temperature, the
material undergoes a p-n transition. At 300K, the thermopower is 20 uV/K. In
an attempt to further probe this transition from p-type to n-type, charge transport

measurements were made on a pressed pellet of the binary or—UTe3, for

299

comparison (see Figure 6.9). The thermopower data of a-UTe3, which is
independent of grain boundaries, gives a behavior more characteristic of that of a
semiconductor and shows at room temperature value of 550 uV/K. The data is
lacking the p-n transition found in CuxUTe3 (x = 0.25 and 0.33). From these
measurements, it is evident that these properties are drastically affected by the
insertion of copper between the layers of 01 —UTe3. Due to the low temperature at
which the p-n transition occurs for CuxUTe3 (x = 0.25 and 0.33), it is difficult to
ascertain the cause. It is interesting, however, that this type of transition has been
reported to occur in MTe5 (M = Zr, lit)30 (at 80K for Hfl‘es and 145K for ZrTes)

and to this date has defied an explanation.

300

 

300

(A) ""

Ill

250

200

V—l—ITTTIUIV‘

150

0' (S/cm)

U11171

100

sot

 

 

 

lJ_llLll l 1111111110441

0 50 100 150 200 250 300 350
Temperature (K)

 

0
lllLLllll

 

 

;.v.l ..n.1.#.1..r.llrgl....lm..'1
0 50 100 150 200 250 300 350
Temperature (K)

 

Figure 6.8 (A) Variable temperature, four probe electrical conductivity for bulk
crystals of CuxUTe3 (x = 0.25 and 0.33). (B) Variable temperature thermopower
data for bulk crystals of CuxUTeg (x = 0.25 and 0.33).

301

 

 

 

 

 

 

 

 

 

 

85 (A)
7;—
e 6
o 5 E
\ 1-
3 4E—
2:-
1 E-
0 EL . s 1 1 . n . l 1 r . r 1 1 14 1 l l 1 r . .
O 50 100 150 200 250 300
Temperature (K)
600 E . o 0‘; .:
L 0 0:“:- 0 .:
550 5 °. . r! 3
500 g W i
A _ i
E 450 3 a
> E 4
1 400 L’ 4:
v i 1
V3 350 E‘ —:
300 :— -§
250 E- . .5
200». n J__A_IA4LJ__L1_J_ mimm n L 14 1 L4
50 100 150 200 250 300
Temperature (K)

Figure 6.9 (A) Variable temperature, four probe electrical conductivity for a

room temperature pressed pellet of (X-UTC3. (B) Variable temperature

thermopower data for a room temperature pressed pellet of (X-UTC3.

302

D. Conclusions
The discovery of CUXUT€3 (x = 0.25 and 0.33) has provided us with the
opportunity to take a closer look at the relative stabilities of the binary UTe3

structure types and has given us some insight as to how the structure of CuxUTeg
(x = 0.25 and 0.33) may be stabilized. As a result, we have determined that 01-
UTe3 is less thermodynamically stable than B-UTe; and that by inserting copper
between the layers of or-UTe3, the physical properties of the material are
drastically affected. Although our attempts to insert Cu directly between the
layers of 111-UTe3 through solid state diffusion methods were unsuccessful, others
have reported on the electrochemical insertion of Cu in ZrTe3,.31 It would be
interesting to see if c0pper could be inserted in 111-UTe3 electrochemically. While
electrical conductivity measurements indicate that CuxUTe3 (x = 0.25 and 0.33) is
a semiconductor, thermopower measurements revealed an interesting p-n
transition at low temperatures. Finally, electron diffraction studies indicate the
existence a 6. 0-6.25a,ub x beub supercell. The type of supercell depends on the
amount of copper in the compound and is assumed to be electronically driven by

structural modulations within the Te chains of the structure.

303

 

References

l

Narducci, A.A.; Ibers, J .A. Chem. Mater. 1998, 10, 2811.

(a) Cody, J.A.; Mansuetto, M.F.; Pell, M.A.; Chien, S.; Ibers, J.A. J Alloys
Compd. 1995, 219, 59. (b) Cody, J.A.; Ibers, J.A. Inorg. Chem. 1995, 34,
3165.

(a) Sutorik, A.C.; Albritton—Thomas, J.; Hogan, T.; Kannewurf, C.R.;
Kanatzidis, M.G. Chem. Mater. 1996, 8 751.

Tougait, 0.; Daoudi, A.; Potel, M.; Noél, H. Mater. Res. Bull. 1997, 32, 1239-
1245.

Chondroudis, K.; Kanatzidis, M.G. C. R. Acad. Sci. Paris 1996, 322, 887-894.
Chondroudis, K.; Kanatzidis, M.G. J Am. Chem. Soc. 1997, 119, 2574-2575.

Sutorik, A.C.; Patschke, R.; Schindler, J .; Kannewurf, C.R.; Kanatzidis, M.G.
Chem. Eur. J, In press.

Choi, K.-S.; Kanatzidis, M.G. Chem. Mater., In press

Daoudi, A.; Lamire, M.; Levet, J .C.; N0el, H. J Solid State Chem. 1996, 123,
331.

‘° (a) Noel, 11.; Potel, M. J Less-Common Met. 1985, 113, 11.15 (b) Noel, H. J

12

l3

l4

Less-Common Met. 1980, 72, 45-49.
Narducci, A.A.; Ibers, J.A. Inorg. Chem, 1998, 379838001.

Sourisseau, C.; Gwet, S.P.; Gard, P. J Solid State Chem. 1988, 72, 257-271
and references therein.

Finckh, W.; Felser, C.; Tremel, W.; Ouvrard, G. J Alloys Comp. 1997, 262,
97-100.

(3) Cody, J.A.; Ibers, J.A. Inorg. Chem, 1996, 35, 3836—3838. (b) Wu, E.J.;
Pell, M.A.; Ibers, J .A. J Alloys. Comp, 1997, 255, 106-109. (c) Choi, K.-S.;
Patschke, R.; Billinge, S.J.L.; Waner, M.J.; Dantus, M.; Kanatzidis, M.G. J
Am. Chem. Soc., 1998, 120, 10706-10714.

304

 

'5 Sheldrick, GM. SHELXTL, Version 5; Siemens Analytical Xray Systems,
Inc.; Madison, WI, 1994.

'6 SMART: Siemens Analytical Xray Systems, Inc., Madison, WI, 1994.

'7 SAINT: Version 4.0, Siemens Analytical Xray Systems, Inc., Madison WI
(1994-1996).

'8 SADABS: Sheldrick, GM. University of Glittengen, Germany, to be
published.

'9 Noel, H.; Levet, J.C. J Solid State Chem. 1989, 79, 28-33.

2° Boehme, D.R.; Nichols, M.; Snyder, R.L.; Matheis, D.P. J. Alloys Compd.,
1992, 179, 37-59.

2‘ Haneveld, A]. Klein; Jellinek, F. J Less-Common Met. 1969, 18, 123-129.

22 (a) Stowe, K. J Solid State Chem, 1996, 127, 202-210. (b) Haneveld, A.J.
Klein; Jellinek, F. J Less-Common Met, 1970, 21, 45-49.

23 Suski, W.; Wojakowski, A.; Blaise, A.; Salmon, P.; Foumier, J.; Mydlarz, T.
J. Magn. Mater. 1976, 3, 195.

2“ (a) Breeze, E.W.; Brett, N.H. J. Nucl. Mater. 1971, 40, 113-115. (b) Breeze,
E.W.; Brett, N.H.; White, J. J Nucl. Mater. 1971, 39, 157-165.

25 Tougait, O.; Potel, M.; Levet, J .C.; Noel, H. Eur. J Sol. State Inor. 1998, 35,
67-76.

2° (a) Noel, H. Inorg. Chim. Acta., 1985, 109, 205207. (b) Noel, H. Mater. Res.
Bull. 1984, 19,1171-1175.

27 (a) Tougait, O.; Potel, M.; Padiou, J.; Noel, H J Alloys Compd. 1997, 262,
320-324. (b) Sttiwe, K. Z.anorg. allg. Chem. 1996, 622, 1423-1427.

2" Tougait, o.; Potel, M.; Noel, H. J. Solid State Chem. 1998, 139, 356-361.

29 Tougait, o.; Potel, M.; Noel, H. Inorg. Chem. 1998,37 (20), 5088-5091.

305

 

3° Littleton, R.T.; Tritt, T.M.; Feger, CR; Kolis, J.; Wilson, M.L.; Marone, M.;
Payne, J.; Verebeli, 1).; Levy, F. Appl. Phys. Lett., 1998, 72 (16), 2056-2058.

3' Finchk, W.; Felser, C.; Tremel, W.; Ouvrard, G. J Alloys Comp. 1997, 262-
263, 97. '

306

Chapter 7
Synthesis and Thermoelectric Studies of the Cage Compounds, AzMCusTew

(A = K, Rb, Cs; M = Ba, Eu)

307

A. Introduction

Recently, there has been an increased interest in finding better materials for
thermoelectric applications.1 As of today, the leading thermoelectric material
remains as BizTe3 and its alloys. Although this material is excellent for use in
small scale thermoelectric devices, it finds its limitation when trying to apply it to
larger scale refi'igeration units. The problem with this material is that it has a poor
efficiency and is therefore very expensive to use. The efficiency of a
thermoelectric material is directly related to a property called the figure of merit.
The figure of merit, ZT, is a unitless parameter made up of the thermOpower (S),
the electrical conductivity (0), and the thermal conductivity (K), see equation 1.

Szo'T
ZT : K (eq. 1)

 

In order to increase the efficiency of a material, one must maximize ZT by
simultaneously maximizing the thermopower (S) and electrical conductivity (6)
while minimizing the thermal conductivity (K). This has proven to be very
challenging since these parameters are not independently controllable. While
many scientists have focused their efforts on optimizing known materials through
doping studies, we have decided to take more of an “exploratory” approach. This
approach has proven, to be very useful and through it, we have found many

promising new materials. One of the most interesting is a family of compounds

308

having the formula AzMCllgTelo (A = K, Rb, Cs; M = Ba, Eu).2 These compounds
fit the description for a “phonon-glass-electron crystal” (PGEC) which was
introduced by Slack3 as the limiting characteristic for a superior thermoelectric.
A PGEC material features cages (or tunnels) in its crystal structure inside which
reside atoms small enough to “rattle”. This “rattling” creates a phonon damping
effect which causes the thermal conductivity to be dramatically reduced. One
system in which this rattling effect has been well illustrated is the skutterudites.4
Here, we report on the anisotropic two-dimensional stucture of A2BaCu3Tem (A =
K, Rb, Cs) and discuss it in relation to its electrical conductivity, thermopower,
and heat capacity. In addition, substitutional doping experiments were performed

associated with the Ba sites.

B. Experimental Section

1. Reagents — The following reagents were used as obtained: Potassium
metal, analytical reagent, Spectrum Chemical Mfg. Corp., Gardena, CA;
Rubidium metal, 99.5%, Alfa Aesar, Ward Hill, MA.; Cesium metal, 99.98%, Alfa
Aesar, Ward Hill, MA; Barium Telluride, 99.5%, 25 mesh, Cerac, Milwaukee,
WI; Europium metal powder, 99.9%, <250 mesh, Alfa Aesar, Ward Hill, MA;
Europium metal chunk, 99.9%, Chinese Rare Earth Information Center, Inner
Mongolia, China; Copper metal, electrolytic dust, Fisher Scientific, Fairlawn, NJ;
Tellurium powder, 100 mesh, 99.95% purity, Aldrich Chemical Co., Milwaukee,

WI. Tellurium shots, 99.9% pure, Noranda Advanced Materials, Saint-Laurent,

309

Quebec, Canada. * The europium metal chunk was cut into fine shavings with a
hacksaw and flamed under vacuum in a sealed quartz ampoule to remove the
oxide coating before being used. The tellurium shots were ground to a fine
powder before being used.

Potassium Telluride, K 2T e - Synthesis of this material was performed as
described in Chapter, Section B. 1.

Rubidium T elluride, szT e - Synthesis of this material was performed as
described in Chapter, Section B. 1.

Cesium T elluride, Cs 2T e - Synthesis of this material was performed as
described in Chapter, Section B. 1.

Europium T elluride, EuT e - The following procedure was modified from
that given in the literature.5 4.330g (0.028 mol) of Eu powder was weighed in an
N; filled glovebox and combined with 3.636g (0.028 mmol) Te in a 500 mL single
neck round bottom flask. The flask was connected to a glass adapter with a
stopcock joint and removed from the glovebox. The flask and adapter was then
connected to a condenser apparatus and chilled to -7 8°C using a dry ice/acetone
bath. Approximately 400mL of NH; were condensed, under an N; atmosphere,
onto the reagents, giving a dark blue solution. The solution was stirred via a
Teflon coated magnetic stir bar and the reaction mixture was maintained at -78°
for up to 24 hours. The dry ice was then removed and the NH; was allowed to
evaporate off as the flask warmed up to room temperature under a constant flow of

N2 (approximately 10 hours). A second portion of NH; was added and the process

310

was repeated to ensure complete reaction of the reagents. The resulting light
brown powder was evacuated on a Schlenck line for approximately 5 hours and
taken into an N2 filled glovebox where it was ground to a fine powder. Due to
europium’s strong ability to coordinate to ammonia, another step was necessary to
remove all of the ammonia from the product. The powder was loaded into a 13
mm silica ampoule and placed on a vacuum line outside of the glovebox. The
ampoule was gently heated with a flame under dynamic vacuum to remove the
coordinated ammonia. The ampoule was then flame sealed and taken back into the
N2 filled glovebox. Caution: If the last step to remove the coordinated ammonia
is not performed, one risks an explosion in subsequent reactions of EuT e in closed
ampoules.

2. Synthesis — All manipulations were carried out under a dry nitrogen
atmosphere in a Vacuum Atmospheres Dri-Iab glovebox.

AzBaCllgT cm (A = K, Rb, Cs) (1-111) and szEuCugT e10 (110 —
Polycrystalline ingots of A2BaCu3Tem (A = K, Rb, Cs) and szEUCUgTelo were
prepared in two steps. A polycrystalline powder was first synthesized by heating a
mixture of AzTe (1 mmol), BaTe or EuTe (1 mmol), Cu (8 mmol), and Te (8
mmol) in a computer controlled fumace to 520°C in 12 hours, isotherming at this
temperature for 4-7 hours, and quickly cooling to room temperature in 5 hours.
The polycrystalline powder sample is then placed in a long silica tube with a
diameter of 5mm and flame sealed. The sealed tube is placed into a gentle flame

to melt the compound and immediately quenched to liquid nitrogen temperatures.

311

This liquid nitrogen quenching process helps avoid the formation of vacuum
pockets inside the ingot. An alternative method to prepare these ingots is to
combine the two steps into one by directly reacting the starting materials in the
flame and quenching the molten product in liquid nitrogen. No isolation was
needed since the compounds were prepared from direct combination of the
elements. The identity of both the polycrystalline powder samples and the ingots
of A2BaCu3Telo (A = K,.Rb, Cs) were confirmed by comparing the powder X-ray
diffraction patterns of the products against the one calculated using single crystal
X-ray data (see Tables 7.1-7.3). The identity of szEUCUgTClo was confirmed by
comparing the powder X-ray diffraction pattern of the product against the

calculated for szBaCugTelo (see Table 7.4)

312

—r1

Table 7.1 Calculated and Observed X-ray Powder Diffraction Pattern for
KzBaCllgTelo (I)

 

 

h kl 9.114) 41.14) III... (obs) 0%)
2 0 0 11.4000 11.4696 71.22
4 0 0 5.7000 5.7412 50.24
2 0 —2 3.5089 3.5211 60.00
2 2 0 3.3240 3.3438 53.17
4 0 —2 3.2541 3.2297 53.17
4 2 0 2.9671 2.9594 100.00
6 0 —2 2.8508 2.8638 71.71
5 1 —2 2.8003 2.8098 59.02
5 3 0 2.0654 2.0696 53.17
5 3 —1 2.0289 2.0214 65.85
0 2 3 1.9229 1.9241 62.44

 

313

Table 7.2 Calculated and Observed X—ray Powder Diffiaction Pattern for
szBaCugTelo (11)

 

 

h kl 9....(4) 45(4) Ill... (obs) 1%)
2 0 0 11.6671 1.6400 57.01
4 0 0 5.8335 5.8643 38.91
5 1 -1 3.7114 3.7172 36.20
2 0 —2 3.5009 3.4718 28.51
4 0 —2 3.3047 3.2873 37.56
1 1 —2 3.1266 3.2223 42.08
6 0 1 3.0745 3.0666 81.45
7 1 —1 3.0063 2.9954 100.00
5 1 2 2.3404 2.3436 25.79
7 1 2 2.0468 2.0450 46.61
7 1 —3 2.0369 2.0395 40.72
8 2 1 2.0245 2.0248 45.25
3 1 3 1.9686 1.9682 34.39
3 3 —2 1.9375 1.9365 56.11
7 3 —1 1.9170 1.9127 47.51
6 2 —3 1.8674 1.8691 38.91
3 3 2 1.8179 1.8207 43.44
5 1 3 1.8092 1.8160 33.94

 

314

Table 7.3 Calculated and Observed X-ray Powder Diffraction Pattern for
CSzBaCUgTCm (HI)

 

 

h kl 9.1.14) 9.1.14) III... (obs) 1%)
0 2 0 11.8805 11.9806 27.74
0 4 0 5.9402 5.9651 17.20
1 4 1 3.8143 3.8137 35.78
2 0 0 3.5545 3.6627 14.01
0 0 2 3.4830 3.4766 23.86
0 2 2 3.3423 3.3331 19.69
2 1 1 3.1384 3.2329 21.08
1 6 1 3.0985 3.0973 66.99
2 4 0 3.0501 3.0507 100.00
2 3 1 2.9399 2.9370 38.97
1 11 0 2.0668 2.0675 26.77
1 9 2 2.0175 2.0187 17.89
0 12 0 1.9801 1.9810 19.00
0 10 2 1.9629 1.9633 19.28
2 1 3 1.9375 1.9367 33.98
3 3 2 1.9019 1.9011 23.44
2 11 1 1.7844 1.7845 29.13
0 0 4 1.7415 1.7405 12.34
0 14 0 1.6972 1.6979 18.72

 

315

 

Table 7.4 Calculated and Observed X-ray Powder Diffiaction Pattern for
szEllCllgTelo (IV)

 

 

h kl 4.1.14) dual) III... (obs) 1%)
2 0 0 11.6671 11.8960 52.96
4 0 0 5.8335 5.8660 34.96
6 0 -1 3.7905 3.7879 30.59
2 0 -2 3.5009 3.4683 32.13
0 0 2 3.4039 3.4069 32.13
2 2 0 3.3694 3.3371 47.81
4 0 -2 3.3047 3.3024 36.76
6 0 1 3.0745 3.0623 81.23
7 1 -1 3.0063 3.0017 100.00
4 2 -1 2.8912 2.8860 70.44
4 0 2 2.6744 2.6984 37.28
6 2 0 2.6095 2.6071 27.51
7 1 1 2.5577 2.5639 43.70
8 0 1 2.4750 2.4634 23.65
9 1 -2 2.2307 2.2322 26.99
6 0 -3 2.2031 2.2070 30.33
5 3 -1 2.0669 2.0741 62.98
7 1 -3 2.0369 2.0387 60.41
8 0 2 1.9917 1.9927 47.56
11 1 -2 1.9523 1.9578 37.02
7 3 -1 1.9170 1.9155 60.41
5 3 -2 1.8774 1.8789 44.47
1 1 -4 1.6779 1.6795 29.31

 

316

3. Physical Measurements — The instrumentation and experimental setup
for the following measurements are the same as described in Chapter 2, Section
3.3: Powder X-ray Diffraction, Infrared Spectroscopy, and Magnetic
Susceptibility Measurements.

Charge Transport Measurements — The thermopower and electrical
conductivity were measured at Clemson University in collaboration with Professor
Terry Tritt and his student, Nathan Lowhom, using two different systems. One
was a helium flow cryostat with a temperature range from 3 - 310 K. The other
was a closed-cycle refrigeration system with a temperature range fiom 12 - 310 K.
Both systems were computer-controlled using LabVIEW software. Samples were
mounted by electroplating the ends of the samples with Ni and then soldering
them to two copper blocks. The two copper blocks served as current leads and
thermoelectric voltage leads. The two junCtions of a 3 mil AuF e (0.07 at. % Fe)
vs. Chrome] thermocouple were embedded in the copper blocks to measure the
temperature difference across the sample. (A small amount of epoxy and “cigarette
paper” was used to electrically insulate the junction.) A 39 Q metal-film resistor
was attached as a heater to one of the copper blocks. Au voltage leads for
measuring resistance were attached to the sample with silver paint. This whole
apparatus attached to a “custom designed chip” that plugged into the measurement
system. One of the copper blocks was thermally connected to the chip which was

thermally sunk to the probe head of the helium flow cryostat or the cold finger of

317

 

the closed cycle refrigerator system. Typical sample size was 3-4 mm in diameter
and 4-7 mm long.

The heat capacity measurements were also performed in collaboration with
Professor Terry Tritt and his student, Nathan Lowhom, at Clemson University.
The PPMS system used has a temperature range of 1.8 - 400 K, can accommodate
samples of mass from 1 - 200 mg, and is fully automated and computer-controlled.
The sample platform has a heater and thennometer underneath and is contained in
a removable sample puck. “Apiezon N grease” was placed on the sample
platform, and the heat capacity of the puck with grease was measured. Then the
sample was placed in the grease for thermal contact to the platform; and the heat
capacity of the puck, grease, and sample was measured. The heat capacity of the
puck with grease was subtracted from the total to give the heat capacity of the
sample. The system uses a relaxation technique to measure the heat capacity.

Raman Spectroscopy - Raman spectra were recorded on a Holoprobe
Raman Spectrograph equipped with a 633 nm HeNe laser and a CCD camera
detector. The instrument was coupled to an Olympus BX60 microscope. Each
sample was simply placed onto a small glass slide and a 50x objective lens was
used to choose the area of the specimen to be measured. The spot size of the laser
beam when using the 50x objective lens was 10pm.

Dlflerential Thermal Analysis (DT A) — DTA experiments were performed
on a computer controlled Shimadzu DTA-50 thermal analyzer. Approximately

30-60mg of sample was ground and placed in a quartz ampoule and flame sealed

318

under vacuum. For a reference, a quartz ampoule of equal mass was filled with
A1203 and flame sealed under vacuum. The sample and reference were heated to
the desired temperature at 10°C/min, isothermed for 10 minutes, and cooled to
50°C at -lO°C/min. The residues of the DTA experiment was examined by power
X-ray diffraction. To determine if the sample has congruently melted, the X-ray
powder pattern before and after the DTA experiments were compared.

Single Crystal X-ray Dlfitaction — The single crystal X-ray data for both
KzBaCugTem (I) and CszBaCugTem (II) were collected previously by Dr. Xiang
Zhang, who also performed the structure solutions.2 Intensity data for
szBaCugTelo (H) were collected on a Siemens SMART Platform CCD
diffractometer using graphite monochromatized Mo K01 radiation. A single crystal
was mounted on the tip of a glass fiber. The data were collected over a full sphere
of reciprocal space, up to 50° in 20. The individual frames were measured with an
0) rotation of 03° and acquisition time of 30 sec/frame. The SMART6 software
was used for the data acquisition and SAINT 7 for the data extraction and
reduction. The absorption correction was performed using SADABSS. The
structure was solved by direct methods using the SHELXTL9 package of
crystallographic programs. Complete data collection parameters and details of all

structure solutions and refinements are given in Tables 7.5.

319

u‘:

”‘2‘“ w

 

 

Table 7.5 Crystallographic Data for A2BaCu3Te10 (A = K, Rb, Cs)
*KzBaCugTem szBaCugTem *CszBaCuchw

crystal habit, color chunk, black chunk, black chunk, black

Diffractometer Rigaku AF C68 Siemens SMART Rigaku AFC6S

Platform CCD

Radiation Mo-K01 (0.71069A) Mo-Ka (0.71073A) Mo-Ka (0.71069A)

Crystal Size, mm3 0.40 x 0.15 x 0.10 0.23 x 0.x28 x 0.05 0.60 x 0.25 x 0.10

Temperature, K 293 173 293

Crystal System monoclinic monoclinic orthorhombic

Space Group C2/m (#12) C2/m (#12) Im (#71)

a, A 23.245(5) 24.034(5) 7.109(2)

b, A 6.950(5) 7.0388(14) 23.761(16)

c, A 7.061(5) 7.0120(14) 6.966(3)

B, ° 101.23(4) 103.86(3) N/A

v, A3 1119(1) 1151.7(4) 1177(2)

Z 2 2 2

11, mm1 22.39 25.509 23.97

indexranges 05h5h -315h530 05h5h
05k5k -95k59 05k5k
-l5l5l -85l59 0515]

20m, deg 50 50 50

sec/flame N/A 30 N/A

total data 3577 5276 658

unique data 3504 I425 634

R(int) N/A 0.0961 N/A

no. parameters 57 57 35

final R/Rw‘, % 8.8/1 1.30 N/A 3.0/6.0

final Rl/sz", % N/A 11.38/30.22 N/A

GOF 4.80 N/A 2.87

Goof N/A 11.47 N/A

 

Rina] -TF,1)/231F,T Rw={Z[W(Fo-Fc)2]/Z[W(Fo)2]}m
l’R1=):(|F,| - |F,|)/2|l=,| wR2={2[w(F..2-Fifi/.53[VlI(I‘..2)2]}"2

* The single crystal X-ray data for KzBaCugTeto and CszBaCusTeto were collected
previously by Dr. Xiang Zhang, who also performed the structure solutions.2

320

 

C. Results and Discussion
Structure Description —- The structure type of AzBaCllgTelo (A = K, Rb,

Cs) has been found to strongly depend on the identity of the alkali metal. While
KzBaCugTelo and szMCUgTClo (M = Ba, Eu) crystallize in the monoclinic space
group, C2/m, CSzBaCllgTCm crystallizes in the orthorhombic space group, Irnmm.
Therefore, all four compounds are not isostructural. However, the two structure
types are very similar. Figure 7.1 shows the extended structure of szBaCugTew.
It is a two-dimensional structure built flom anionic [BaCugTe10]2’ layers that are
separated by Rb+ cations. The [BaCugTelojz' layers can be further broken down
into CugTeu pentagonal dodecahedral cages, see Figure 7.2A. These cages each
encapsalate one Ba2+ cation, which are coordinated by twelve Te atoms.
Interestingly, the cages have a strong affinity for cations with a high charge/radius
ratio, and so, given a choice between a Rb” and Ba2+ cation, it will encapalate the
latter. This preference appears to be electrostatic in origin so as to reduce the
negative charge of the [CugTe10]4‘ cage. The coordination environment of the
copper atoms is a slightly distorted tetrahedron. Each CllgTCjz cage contains three
mutually perpendicular sets of ditelluride units. By sharing two pairs of
oppositely spaced ditellurides, the cages form layers along the a-axis. The third
ditelluride unit remains unshared. In addition to the three ditelluride units, each
cage possesses monotellurides which are coordinated to four Cu atoms in a square
pyramidal geometry. The Te-Te bond distance of the three ditelluride units range

flom 2.798(5) A to 2.862(5) A, while the Cu-Te bond distances range flom

321

 

 

2.576(4) A to 2.686(5) A. The Rb+ cations are stabilized between the layers and
are each coordinated by ten Te atoms. If the Cu atoms are included, one can see
that the cations are actually sitting inside a “cup” that is made up of part of a
pentagonal dodecahedral cage, see Figure 7.28. The flactional atomic
coordinates, isotropic and anisotropic temperature factors, bond distances, and
bond angles for szBaCugTelo is given in Tables 7.6-7.8.

The extended structure of CszBaCugTem is shown in Figure 7.3. The
difference between this structure type and that of szBaCugTelo lies in the way
that the anionic layers stack with respect to one another. In szBaCugTem, the RbJr
ions are each coordinated by ten Te atoms. While this coordination environment
is large enough to stabilize potassium and rubidium, it is not large to stabilize a
cesium ion. Therefore, the [BaCugTem]2' layers are forced to shift with respect to
one another to create a larger space for the cesium ion to reside. As a result, the
symmetry of the structure drops flom orthorhombic to monoclinic. As shown in
Figure 7.4B, the cesium ions are now coordinated to eleven Te atoms in a tri-
capped square antiprismatic geometry and again sit inside a “cup” formed flom

Part of a pentagonal dodecahedral cage.

322

 

°Rb

      

4 ° .1
«34:8»

.mp0

 

Figure 7.1 ORTEP representation of the extended structure of szBflCUgTClo as
seen down the b-axis (90% ellipsoid probability). The ellipses with octant shading
represent Ba atoms, the crossed ellipses represent Cu atoms, and the large open

ellipses represent Rb and Te atoms.

323

 

 

 

 

 

Figure 7.2 (A) ORTEP representation of the barium filled [ClixTCn] cages of
szBaCugTem (50% ellipsoid probability ellipsoid) and (B) the coordination
environment around Rb in szBflCUgTClo. Bond distances include: Rb-Tel =
3.848(6)A, Rb-Te3‘ = 4.057(7)A, Rb-Te3" = 3.729(5)A, Rb-Te4a = 3.742(2)A,
Rb-Te4b = 3.906(8) A. The ellipse with octant shading represents Ba, the crossed

and striped ellipses represent Cu atoms, and the large open ellipses represent Te

atoms.

324

 

9 O
0.- o

—> c"
‘C’
. .

 

 

 

Figure 7.3 ORTEP representation of the extended structure of CszBaCugTelo as
seen down the a-axis (90% probability ellipsoids). The ellipses with octant shading

I‘Cpresent Ba atoms, the crossed ellipses represent Cu atoms, and the large open

ellipses represent Cs and Te atoms.

325

 

 

Figure 7.4 Coordination environments around (A) Rb in szBaCugTem (for

comparison) and (B) Cs in CszBaCugTem. Bond distances include: Cs-Telb =
3.959 A, Cs-Te3° = 4.238 A, Cs-Te3b = 3.825 A, Cs-Te4a = 3.820 A, Cs-Te4 =

4.328 A

326

 

 

 

____.,.._. ..44 w..-

 

Table 7.6 Fractional Atomic Coordinates and Equivalent Isotropic Displacement
Parameters (Ueq) for szBaCugTem with Estimated Standard Deviations in

Parentheses.

 

 

atom x y z occupancy Ueq’, A2
Ba 0.0 0.50 0.0 1.0 2.0(1)
Te(l) 0.06000) 0.50 0.4518(4) 1.0 1.6(1)
Te(2) 0.0030(1) 0.0 0.2043(3) 1.0 1.5(1)
Te(3) 0.14580) 0.3031 0.1190(3) 1.0 1.9(1)
Te(4) 0.1707(1) 0.0 0.6413(4) 1.0 1.8(1)
Cu(l) 0.0949(2) 0.1920(5) 0.3843(6) 1.0 2.4(1)
Cu(2) 0.0940(2) 0.1930(5) 0.2286(5) 1.0 22(1)
Rb 0.2247(2) 0.50 0.707603) 1.0 55(3)

 

“Ueq is defined as one-third of the trace of the orthogonalized Uij tensor.

Table 7.7 Anisotropic Displacement Parameters (A) for szBaCugTelo with
Standard Deviations in Parentheses.

 

 

atom U11 U22 U33 U23 U13 012
Ba 0.029(2) 0.013(2) 0.021(2) 0 0.014(1) 0
Te(l) 0.026(1) 0.007(1) 0.018(1) 0 0.014(1) 0
Te(2) 0.025(1) 0.008(1) 0.015(1) 0 0.012(1) 0
Te(3) 0.029(1) 0.012(1) 0.020(1) 0.001(1) 0.014(1) 0.003(1)
Te(4) 0.028(1) 0.011(1) 0.019(1) 0 0.016(1) 0
Cu(l) 0.042(2) 0.013(2) 0.022(2) -0.002(1) 0.015(2) -0.002(1)
Cu(2) 0.036(2) 0.014(2) 0.021(2) -0.003(1) 0.017(1) 0.001(1)
Rb 0.029(3) 0.023(2) 0.1 15(6) 0 0.021(3) 0

 

327

 

 

Table 7.8

Selected Distances (A) and Bond Angles (deg) for szBaCugTem
with Standard Deviations in Parentheses.

 

 

 

 

Bond Distances

Cul— Tel 2.681(4) Cu2 — Cu2 2.717(7)

Cul — Te2 2.686(5) Tel — Tel 2.798(5)

Cul — Te3 2.581(4) Te2 — Te2 2.862(5)

Cul - Te4 2.612(5) Te3 — Te3 2.772(4)

Cu2 - Tel 2.679(4) Rb — Tel 3.848(6)

Cu2 — Te2 2.681(4) Rb — Te3 3.729(5), 4.057(7), 4.336(8)

Cu2 — Te3 2.576(4) Rb — Te4 3.742(2), 3.906(8)

Cu2 — Te4 2.623(4) Ba — Tel 3.768(3)

Cu] — Cu] 2.703(7) Ba — Te2 3.7791(11)

Cul - Cu2 2.720(5) Ba — Te3 3.673(2)

Bond Angles

Tel -— Cul — Te2 106.9706) Te2 - Cu2 — Te3 108.4504)

Tel — Cul - Te3 108.3704) Te2 — Cu2 - Te4 115.74(l4)

Tel - Cul — Te4 111.1505) Te3 — Cu2 — Te4 105.33(16)

Te2 — Cul — Te3 107.9405) Tel — Rb — Te3 65.3500), 144.61(6)

Te2 — Cul - Te4 115.8404) Tel — Rb - Te4 7022(9), 126.2(2)

Te3 — Cul — Te4 106.33(l6) Te3 — Rb — Te3 3995(9), 69.8002),
93.9405), 1 17.4(2)

Tel - Cu2 — Te2 107.5506) Tel — Ba - Tel 180.0

Tel -— Cu2 — Te3 108.7204) Tel — Ba — Te3 69.96(5),l 1004(5)

Tel —— Cu2 —- Te4 110.8504) Te3 — Ba —- Te3 4433(6), 135.67(6),

180.0

 

328

Charge Transport Properties - Physical measurements were previously
made on single crystals of szBaCusTelo and the results are shown in Figure 7.5.
The electrical conductivity mostly decreases with decreasing temperature which is
characteristic of a semiconductor. Above 200K, however, the conductivity
flattens out to a value around 125 pV/k and then starts to decrease around 250K.
This is an indication that the bandgap of this material is very small and that at
higher temperatures, the material acts more like a semimetal. The thermopower
data is given in Figure 7.53 and suggests p—type behavior. Although the
magnitude of the thermopower suggest semiconducting behavior, the slope is
more reminiscent of a metal. Besides the anomalies in the data, the results
seemed very promising for thermoelectrics, especially when taking into
consideration the very low thermal conductivity. The thermal conductivity (now
shown here) is 3.5 mW/cm-K at room temperature and ranges from 2.5-5.0 above
180K. This coupled with the fact that these compounds melt congruently, which
is also a very important property for thermoelectrics, led us to perform an
extensive study on these materials.

Figures 7 .6-7 .8 show both the electrical conductivity and thennopower for
polycrystalline ingots of AzBaCugTelo (A = K, Rb, Cs), respectively. For all three
materials, the electrical conductivity data give metallic behavior with room
temperatures ranging from 28 - 580 S/cm. The thermopower values range from
approximately 30—55 pV/K at room temperature, suggesting p-type behavior. The

common feature of all the thermOpower data is that it decreases with decreasing

329

 

temperature until exhibiting a small peak below 50K. This peak is believed to be a
phonon drag peak as magneto-heat capacity measurements reveal no magnetic
effects in these samples.

By comparing these measurements to those of the single crystal
measurements for szBaCugTe 10, a few things become apparent. First, the single
crystal data suggest the material to be semiconductor while those on the ingots
suggest metallic behavior. This difference cannot simply be explained by grain
boundary effects in the ingot because this effect acts to inhibit the conductivity,
not enhance it. This suggests that perhaps the materials are doped and that the
doping impurities act to create a degenerate semiconductor. This is supported by
the fact that the magnitudes of the conductivity have such a wide range. Although
not discussed here, there exists two ternary compounds that are isostructural to
AzBaCugTem (A = K, Rb, Cs). These are Rb3CUsTClo and CS3CU3TC]0.2 The
difference is that for these compounds, the Rb)r and Cs + ions also occupy the space
inside the cages of the structure. Charge transport measurements on these ternary
compounds indicate p-type metallic behavior. Knowing this, one possible
explanation is that the samples are actually a solid solution between AzBaCugTelo
and A3Cu3Te10, To test for this, polycrystalline ingots were made with an excess
of Ba (10% - 40%). The excess Ba in the reaction should force the equilibrium

more completely towards the formation of szBflCUgTelo, see Scheme 1.

330

 

Scheme 1.
+ Ba + Ba
Rb3C113T610 Z—a Rb2[Bal-x RdellgTelo —" szBaCllgTem
-Ba -Ba

The results from these experiments are shown in Figure 7.9. The added Ba in the
syntheses appears to have simultaneously decreased the conductivity while
increasing the thermopower in a systematic fashion. The addition of 40% Ba
changes the electrical conductivity fi'om that of a metal to that of a semiconductor.
This data is much closer to that of the single crystal data and therefore supports the
above argument.

Finally, measurements were made on samples where the Ba has been
substituted with Eu. By placing a heavier atom inside the cage, it might be
possible to further minimize the thermal conductivity. Experiments were also
performed to synthesize szEUCU3Telo with an added amount of Eu in the reaction
and measure their properties. The results are shown in Figure 7.10. The
substitution of Eu for Ba in szBaCllgTClo increased the conductivity to
approximately 1800 S/cm at room temperature and decreased the thermopower to
17 pV/K. Additonal Eu in the synthesis had relatively no affect on the properties.

For all measurements, the room temperature values are given in Table 7.9.

331

 

 

 

 

 

 

 

 

 

Electrical Conductivity
140
KM
p
120 _-
a I
g 100 '-
w I
v
b 80 :-
60 -_/
40 IHILJIALAIIIIlJIIJI11.111:-
0 50 100 150 200 250 300
Temperature (K)
Thermopower
F I
503) 5
150 :- M '5
g 1:: §’f~ g
> 100 :- -j
1 b I
v 1; I
m a a
50 '- é
: . i
E . .../ 1
0 {ILLIJJLLLI .14J L4+LLLLnn l 11L. L:
0 50 100 150 200 250 300
Temperature (K)
Figure 7.5 (A) Variable temperature electrical conductivity data for a single

crystal of szBaCllgTCm and (B) Variable temperature thermopower
data for a single crystal of szBaCugTelo.

332

Electrical Conductivity

 

 

 

 

 

 

 

 

 

1400. ...............
1200:-\ (A):
10005-8. .:
€800:- 1' E
2 ; 3 .
@600' =
b : E
400:- -:
200:" fi
0: lI...ll-Ijlljllll-lll-lllll.v
0 50 100 150 200 250 300
Temperature (K)
Thermopower
35 ..,,.
:(B) .
307 .:
g”:- ':
>20:- “I
m15E- '.
10:- '5
5:..IW11
0 50 100 150 200 250 300

Temperature (K)

Figure 7.6 (A) Variable temperature electrical conductivity data for an ingot of
KzBaCugTem and (B) Variable temperature thermopower data for an ingot of

KzBaCllgTCm.

333

 

Electrical Conductivity

 

 

 

 

 

 

 

 

 

 

100000 . . . . . . . .
(A)
A 10000
E
8
re
b 1000 81,83
60 s2
'3 85
100
S4
10_ lllllllIn-alln-n
0 50 100 150 200 250 300
Temperature (K)
Thermopower
70 I I I I I I I I l I I I I I I I I I I I I I I ' I I I I SS
B
60 ( )
50
Q
40
3.
v 30
m
20
10 . .,/
Olllll llll llll nan. 300
0 50 100 150 200 250

Temperature (K)

Figure 7.7 (A) Variable temperature electrical conductivity data and (B)
Variable temperature thermopower data for five ingots of szBaCugTelo (Sammes

Sl- SS).

334 '

Electrical Conductivity

 

 

 

 

 

 

 

,......
. , (A);
\.. .
E “:3 2.x
0 . "s
@1000 \
b -
8"
1—1 88
S9
86,87
100 ...I....1....|....|....1...._
0 50 100 150 200 250 300
Temperature(K)
Thermopower
_IITI'IIII'IIII'IIII'IIII'III 86.89
35 (B) / S7
30 88
A 25
§ 20 /
:l.
V15
:13
10
5 ./\.
0:: llIL.ILlllllJllJlllllllllllf
0 50 100 150 200 250 300

Temperature (K)

Figure 7.8 (A) Variable temperature electrical conductivity data and (B)
Variable temperature thermopower data for four ingots of C82BaCu3Tem (Samples

86 - 89).

335

 

 

 

 

 

 

 

 

Electrical Conductivity
100000 . 70
(A) ,1
1 65
A (d) 1
E 10000 - q
3 . - 60 A
Z” : (4
b ‘ o
60 1 55 3
3 1000 3
~ (6) 4
~— (C) .
100 ialLa LLllLllLlj—Ll nmlh-HI 1_Ll 1L. .4. 45
0 50 100 150 200 250 300 350
Temperature (K)
Thermopower
140 ..
120 (B) /~ ((0 -j
100 4r" .'
g a!
(0) -'
> 80 :
3- 60 (b) .:
w a
20 ..
O Y jam IJLLIJJLI [JLLIJJLIIJJILIJJJLIE

0 50 100 150 200 250 300 350
Temperature (K)

Figure 7.9 (A) Variable temperature electrical conductivity data and (B)
variable temperature thermopower data for (a) szBaCuchlo, (b) szBaCugTem +
0.lBa, (c) szBaCugTelo + 0.3Ba, and (d) szBaCugTelo +0.4Ba. All

measurements were made on ingots.

336

 

 

Electrical Conductivity

 

(A)

Log 0 (S/cm)

 

 

 

 

0 50 100 150 200 250 300 350
Temperature (K)

25 Thermopower

 

 

 

0 50 100 150 200 250 300 350
Temperature (K)

Figure 7.10 (A) Variable temperature electrical conductivity data for (a) a
pressed pellet of szEuCugTem, (b) an ingot of szEllCllgTClo, and (c) an ingot of
szEllCllgTem + 0.2Eu. (B) Variable temperature thermopower data for (b) an
ingot of szEUCUgTClo, and (c) an ingot of szEUCUgTCm + 0.2Eu.

337

 

 

Infrared Spectroscopy - The diffuse reflectance optical spectra was
taken in the Mid-IR region for szBaCugTem, CszBaCugTelo and szEllCllgTem,
see Figure 7.11. Optical gaps are observed at 0.28 eV, 0.30 eV, and 0.23 eV,
respectively, for the three compounds. This data suggest that these materials are
probably semiconductors, despite the fact that the charge transport measurements
suggest metallic behavior. The absorption edge of szEUCllchlo, however, is
much broader than that of AzBaCugTelo (A = Rb, Cs). One explanation for this
might be that this sample has more of a contribution fi'om the ternary Rb3Cu3Tem
compound which makes the overall material more of a semimetal.

Contrary to this data, Density Functional Theory (DFT) calculations have
been performed on CszBaCugTew'o, which gave a calcuated DOS where the Fermi
Level falls into a deep pseudo-gap. Because this gap is not completely empty, the
material must be characterized as semimetal. The narrow- gap semiconductor
versus semimetal characterizaion is hard to address unequivocally at this stage
because we are pushing against the accuracy limits of our computational
technique. Therefore, additional experiments were necessary to resolve this issue,

see heat capacity results below.

338

 

 

 

 

 

  

 

 

 

 

 

 

 

’ ' ' ' I rff f I ' ' r ' I ' ' r ' I ‘ ' ' ' T r ' ‘
r(A) 1
A P i
(I) ' 1
a; 1
LE 1
33 :
v u
m 1
3 4
1
.--l-4-.l---.l..g-l--..l-..-'
0.30 0.40 0.50 0.60 0.70 0.80
Energy (eV)
08 (B)
.fl. _
a.
:3
E‘ .‘I‘
a: I- " .o
g
5,;
m
B.
0.30 0.40 0.50 0.60 0.70 0.80
Energy (eV)
((3) :Iarizé? -1
‘o . $33,344. :
A - - Ff . "' "'
3 .. .rfix-‘Qt-‘E 1
5 ‘ «were: -;
5‘ ‘. 0 O. ..0 g :
.2 E
..o -
2a :
v I
w 3
B .-

 

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Energy (eV)

Figure 7.11 Diffuse reflectance optical spectra of (A) szBaCllchlo, (B)
CszBaCu3Tew and (C) szEUCUgTClo (in the Mid-IR region).

339

Heat Capacity - The heat capacity was measured for both szBaCugTe 10
and CszBaCugTem and the results are shown in Figure 7.12. All samples show a
typical heat capacity temperature dependence. By plotting the data as Heat
capacity/1‘ vs T2, see Figure 7.13, it is possible to derive the Debye temperatures,

OD, through the following relationships,

Cp = YT + 5T3 (eq. 2)
1/3 '
(9D: [4495—] x 10 (eq. 3)

where the first term of eq. 2, 7T, is the electronic contribution to the heat capacity
and the second term, 0T3, is the lattice contribution to the heat capacity. The
Debye temperature is proportional to the stiffiless of the lattice, i.e., the stiffness of
the interatomic force constants and bonding energies. If the Debye temperature is
low, the material should possess a low thermal conductivity. The advantage to
measuring the heat capacity over thermal conductivity is that much less sample is
needed to perform the experiment and the experiment itself is easier to perform.
The Debye temperatures for the szBaCugTelo ranged from 345-3 54K while those
for the CszBaCugTelo samples were somewhat lower ranging from 312-320K, see
Table 7.9. Application of a magnetic field had no effect on the temperature
dependence or the magnitude of the heat capacity. The 7 values for both
compounds were also derived from the low temperature data (Table 7.9) and they

average 0.279 mJ/mol-Kz. For semiconductors with a well-defined energy gap, 7

340

= 0.. Typical metals have yvalues ranging from 0.008 — 10.7 mJ/mol-Kz.
However, there are examples of semiconductors that have a y > 0 within the range
observed here. 11 The deviation fiom zero can be explained again by the argument
that these samples are doped with impurities making the material a degenerate
semiconductor. Therefore, fiom these results, we can conclude that the

AzBaCugTelo (A = Rb, Cs) materials are narrow gap semiconductors.

341

 

 

 

35 Ii I I I I1 I I r1 I—I I T—I I IT I I I—I I I "f I I I
A
:4. 30 (A)
a v a
g 25 ., '0 .1. .8 0; .18 r! .......... E.
E 20 J 0'8 1' 4
g 15 64* -:

fl
8 10 g -j
H d
8 5 1:
m 0 um.LUL.4.JL.l.JL.J..L.4L.JUL '
0 50 100 150 200 250 300 350
Temperature (K)
30

 

rIIT IIIT IfII 'IrII IIII rrII IrII
I l l 1 T T

(B)

     

i

25 " J
1; '° 1

20 -»' " ' 3
.m..[, i

10

Heat Capacity (J /mol-K)

malnwla

 

 

0 jnnlijJLLILLlLAJJJJLLLlAJJnILL44
0 50 100 150 200 250 300 350
Temperature (K)

Figure 7.12 Heat capacity (J/mol-K) data for (A) four ingots of szBflCllgTClo
and (B) three ingots of CszBaCusTem as a function of temperature.

342

 

.°
0.)
O

 

I Ifi 'fil ITI I I r. I—I ' I I Ij Iii

(A) 5

l

p
N
Ut

.O
N
O
DDI
a.

I
find
O

8
Ill-LIIIUIIIIJLIIILIII

 

Heat Capacity/T (J/mol-Kz)
O O
o G

 

j l LJ I LLJ L44 Ll I IJ L4 I L] l I

p
8

 

 

 

 

0 210‘ 4104 6104 8104 l 105 1.2105
2 2
Temperature (K )
NA 0.30 f I I T I —I r I I r i l I T r I r r I I r I I 1
if. (13) a
o 0.25 ._ 1
a q, a
c 0.20 a! a.
b *-. E
.Q 015 ..- 1
O ‘= . 1
C6 - ,
g 010 ., .. 1
Q) .
m 000 I I J J J I I J I J I I 4 I j l j J I l I I I I
0 2104 410‘ 6104 810‘ 1 10’ 1.2105
Temperature 2(K2)

Figure 7.13 Heat capacity/T data (J/mol-Kz) vs T2 data for (A) four ingots of
szBaCugTem and (B) three ingots of CszBaCugTem.

343

 

Table 7.9 Room temperature values for the electrical conductivity,
thermopower, and heat capacity of AzMCUgTClo (A = K, Rb, C8; M = Ba, Eu) and

the Debye temperatures and 7 values derived from the heat capacity

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Compound Type Electrical Thermopower Heat Debye 1 value
Conductivity S (“V/K) Capacity Temperature mJ/ ol
0' (S/cm) (JImol-K) 9., (K) ( m
_K2)
KzBaCugTelo ingot 185 S/cm 33 uV/K ————-— ————— -—-—
szBaCUchlo ingot (81) 588 S/cm 33 uV/K 22 J/mol-K 345K 0.223
szBaCugTem ingot ($2) 233 S/cm 54 uV/K 29 J/mol-K 354K ——
szBaCugTew ingot (S3) 588 S/cm 32 uV/K 27 J/mol-K 350K 0.519
szBaCUgTCw ingot (S4) 28 S/cm 55 uV/K 23 J/mol-K 352K 0.200
szBaCusTew ingot (SS) 120 S/cm 68 uV/K ———--— -—-—-—-— -—-———
szBaCugTem ingot 303 S/cm 59 uV/K -—-—-— —— ~—-——
+ 0.1 Ba
szBaCugTem ingot 182 S/cm 75 uV/K —-——-- --——— -—
+ 0.3 Ba
szBaCugTem ingot 64 S/cm 122 pV/K —————~ ————— -——-—
+ 0.4 Ba
CSzBaCUgTCm ingot (86) 172 S/cm 42 pV/K 24 J/mol-K 320K 0.335
CszBaCugTelo ingot (S7) 179 S/cm 33 pV/K 26 J/mol-K 312K 0.239
CszBaCugTem ingot (S8) 417 S/cm 30 pV/K 25 J/mol-K 312K 0.156
CszBaCusTem ingot (S9) 250 S/cm 40 pV/K ~—-—-—- --—-—-—- ——-——-
szEUCUgTClo pellet 2273 S/cm 14 uV/K -——-—- —-—--— -———
szEUCUgTelo iflgOt 1852 S/cm l7 [IV/K -——————
szEuCugTem ingot 2273 S/cm 20 uV/K __
+ 0.2Eu

 

 

 

 

 

 

 

344

 

 

Raman Spectroscopy - The Raman spectra of szBaCugTem and
CszBaCUgTelo are shown in Figure 7.14 and shows shifts at 121cm" and 140 cm".
These can be assigned as the stretching vibration of the ditelluride units in the
structure. Unfortunately, the experimental apparatus prevented being able to
observe any vibrations below 90cm", which would most likely be the “rattling”

modes of the A+ (A = Rb, C8) or Ba2+ ions.

 

I IrI—II'Ij—I—ITI I I rirrfT1—Ij rrrII1I rrII I

121 cm"1

  
  
   
   

140 cm1

 

Rb 2BaCu 8Tel 0
n 121 cm"

 

140 cm"

  

 

Intensity (arbitrary units)
t

 

 

CszBaCusTem

JLJ—LILJLJIIILILJLJ4I[IILJILJLILILII

100 150 200 250 300 350 400 450
. -1
Raman Shut (cm )

Figure 7.14 Raman Spectra of CszBaCugTelo and szBaCugTelo

345

 

Magnetic Susceptibility -- Variable temperature magnetic susceptibility data
for szEuCusTem was measured over the range of 5-300K at 60006. A plot of
llxM vs T (see Figure 7.15A) shows that this material exhibits perfect Curie-Weiss
behavior. A pcgvalue of 6.95 BM and a Weiss constant of —17K was estimated by
fitting a straight line to the data above 100K. This value is less than what is
expected for an 1‘7 configuration or Eu2+ (7.9 - 8.0 BM) but very different from
that expected for Eu3+ (3.3-3.5 BM).'2 Recall, however, that the charge transport
measurements on szEUCllgTelo indicate that the material is actually a solid
solution between szEuCugTem and Rb3Cu8Tem and the true composition is
Rb2[Eu1.xbe]Cu3Tem. Thus, we should expect a lower peg value since there is
less than one paramagnetic center per formula. That is indeed what is observed
and the data appears to support this argument. The peg value can therefore be used
to calculate the true chemical formula, which is Rb2,24Euo,76CugTelo.

The magnetic susceptibility was also measured for szEllCllgTem at 30006 and
careful diamagnetic corrections were made.

The molar magnetic susceptibility plotted vs temperature for
szBaCugTelo is shown in Figure 7.158. Since there are no unpaired electrons on
the compound, it is expected to behave diamagnetic. However, if the material is
slightly paramagnetic after all diamagnetic corrections are performed, the material
is said to possess Pauli paramagnetism. Pauli paramagnetism is proportional to

the density of states (DOS) at the Fermi Level and therefore, its contribution is due

346

 

to conduction electrons of a metallic material since a semiconducting material has
no free electrons at the Fermi Level.‘l Therefore, it is possible to get a qualitative
idea as to whether or not szBaCugTew is a metal or a semiconductor fiorn the
magnetic susceptibility data. The data shows negative values down to 10K,
suggesting “true” diamagnetic behavior. The positive values below 10K can be
attributed to small paramagnetic impurities in the sample. From this, we can
conclude that there are essentially no free electrons at the Fermi Level and the

material is a narrow gap semiconductor.

347

 

 

 

 

 

 

 

60 _
1 A
50 __( ) .
' e
i o
40 7 o
" O
s? 5 '
\ 30 L" o
v—t . .
3 o
20 '_' . .
' e
I . 0
10 f
OfiIJJ LJJ LLJILJ LLJILJJ Lil-317
0 50 100 150 200 250 300
Temperature (K)
0.0003 .101T1Tr1ri ,1T1 rtj '1 .4 "T‘TTT
B)
0.0002 :1-
0.0001 :-
C
x2 0.0000 [-
00001 E-
;
00002 E-
)-
_0.0003 :ILIIILIIILLIIIJIILL LLL L +
0 50 100 150 200 250 300
Temperature (K)

Figure 7.15 (A) Inverse molar magnetic susceptibility (l/xM) plotted against
temperature (2-300K) for szEUCUgTClo and (B) Molar magnetic susceptibility
(xM) plotted against temperature (2-300K) for szBaCu3Tew.

348

Thermal Analysis - The DTA spectra of AZMCugTem (A = Rb, Cs; M = Ba,
Eu) show several endothermic and exothermic peals below 650°C , see Figure
7.16. The powder X—ray diffraction patterns before and after heating are the same,
suggesting that the materials melt congruently. However, the fact there are several
peaks in the DTA spectra indicates that either the thermal behavior is very
complicated or the materials are not pure. The question of purity can once again
be explained by the previous argument that the material is likely a solid solution
between A3CU3T610 and AzMCllgTelo. While Rb3Cu3Telo has been reported to
decompose around 400°C, szBaCugTelo is believed to melt congruently at
approximately 500°C. From Figure 7.17A, it can therefore be crudely estimated
that the most intense thermal events at 491°C and 454°C are the melting and
recrystallization temperatures, respectively, for szBaCugTem. The other six
thermal events can be attributed to either the melting and recrystallization of the
solid solution phase(s), Rb2+xBa1.xCugTelo (0 < x < 1), or the decomposition of
Rb3Cu3Te10. Figures 7.17B and 7.17C show similar phenomena for szEUCUgTelo
and CszBaCugTem, respectively. From the charge transport measurements, it
could be predicted that, with added Ba or Eu in the synthesis, the products would
be more of a single phase and that the lower temperature thermal events could be
eliminated. The DTA spectra of szMCUgTelo + 0.4M (M = Ba, Eu), however,
suggest otherwise. There does not seem to be any noticeable improvement in the
DTA diagram and it appears as if the location and intensity of the thermal events

varies from sample to sample. In fact, if the intensity of the higher temperature

349

 

thermal events gives any indication as to the percentage of AzMCUgTCm (A = K,
Rb, Cs; M = Ba, Eu) present in the sample over A3CugTem or Rb2+xBa1-xCugTe10
(0 < x < 1), the spectra shown in Figure 7.16 are thus far the most single phase.

Given the observed behavior, we cannot at this stage unequivocally conclude that

these materials actually melt congruently.

350

 

305C 315C 375C
/

f,

 

267C 307C 350C

 

 

 

 

 

 

 

 

 

0 100 200 300 400 500 600 700
Temperature (C)
50 {Ij111f1 I rrI rI—I' rrr I rrt I I I l' I I—rrrrrrra
40 :03) 344C -3
30 IE- .1
> 20 5- .3
1 : 3
10 E- .9
5 E
0 E 'i
-16 'r a
_20 ..LLlsssleLstsmle.”sslssl.1.,141
0 100 200 300 400 500 600 700
Temperature (C)
40 (6)111 r rrIjj TI’ I—I—I rI I ' rIT—I I T—I rf' I Ii ‘1
p :1
30 ? 464C 1
~ 1
5 1
20 :- .1
> - 1
1 I :
10 L- ..
~ 2
0 E- .3
I I
,

 

 

 

10 JILJILILLIJLJ LILLII]_I_I_I_I_J_IJ_LJ_1_I_III
-

0 100 200 300 400 500 600 700
Temperature (C)

Figure 7.16 — DTA diagrams of (A) szBaCugTelo, (B) szEUCllgTClo, and (C)

CSzBaCUgTC 1 o

351

D. Conclusions

Although a lot of work has been done to try and understand the physical
properties of the compounds, AzMCllgTClo (A = K, Rb, Cs; M = Ba, Eu), a lot
more work needs to be done. For example, we are still not completely certain that
the compound szEllCllgTelo really exists. While the physical data collected thus
far seems to suggest that it does, we have not yet been able to obtain single
crystals for a structure determination. Also, we have not yet been able to
synthesize the compounds AzBaCugTelo (A = K, Rb, C8) in a completely pure
ingot form. The “Ba rich” experiments, however, have led us in the right direction
and allowed us to understand that these materials are most likely solid solutions of
the type A2[Ba1-xAx]Cu3Telo (A = K, Rb, Cs), giving rise to semimetallic to
metallic properties. Although we have experimentally observed a bandgap for
these materials, the DF T calculations suggest metallic behavior. By measuring the
heat capacity and measuring the magnetic susceptibility, however, we were able to
support the fact that that these materials are indeed semiconductors and now
believe that the theoretical calculations have underestimated the magnitude of the
gap. Differential thermal analysis appears to be a very useful tool in determining
the purity of the sample. This, coupled with electrical conductivity data should
make it possible to fine-tune the synthesis to yield pure AzBaCllgTC 10 (A = K, Rb,
Cs) samples in ingot form. Once the synthesis is optimized, thermal conductivity

measurements need to be performed. Further optimization of the electrical

352

 

properties will include doping the sample with Se and/or Sb. These experiments

are currently in progress for KzBaCugTelo.

353

 

References

 

12

“Thermoelectric Materials —- New Directions and Approaches”, Mat. Res.
Soc. Symp. Proc., 1997, vol 478. Edited by T.M. Tritt, M.G. Kanatzidis,
H.B. Lyon, and GD. Mahan

Zhang, X.; Park, Y.; Hogan, T.; Schindler, J.L.; Kannewurf, CR; Seong,
S.; Albright, T.; Kanatzidis, M.G. J. Am. Chem. Soc. 1995, 117, 10300.

(a) Slack, G.A.; “New Materials and Performance Limits for Thermoelctric
Cooling” CRC Handbook of Thermoelectrics, CRC Press (Boca Raton,
1995), p 407. (b) Slack G.A., in “Solid State Physics”, eds. Ehmereich, H;
Seitz, F.; Tumbull, D. (Academic, New York, 1977), Vol 34, R].

(a) Nolas, G.S.; Slack, G.A.; Morelli, D.T.; Tritt, T.M., Ehrlich, A.C. J
Appl. Phys. 1996, 79, 4002. (b) Nolas, G.S.; Morelli, D.T.; Tritt, T.M.
Annu. Rev. of Mat. Sci. 1999, 29, 89. ‘3
Parkin, I.P.; Fitzrnaurice, J.C. Polyhedron 1993, 12, 1569.

SMART: Siemens Analytical Xray Systems, Inc., Madison, WI, 1994.

SAINT: Version 4.0, Siemens Analytical Xray Systems, Inc., Madison WI,
1994-1996.

 

 

 

SADABS: Sheldrick, GM. University of G6ttengen, Germany, to be
published.

Sheldrick, GM. SHELXTL, Version 5; Siemens Analytical Xray Systems,
Inc.; Madison, WI, 1994.

Patschke, R.; Zhang, X.; Singh, D.; Schindler, J .; Kannewurf, C.R.;
Lowhorn, N.; Tritt, T.; Nolas, G.; Kanatzidis, M.G. Manuscript in
preparation.

Kittel, C., Introduction to Solid State Physics, John Wiley & Sons, Inc.
1986, pp 141 and 413-416.

Greenwood, N. N.; Eamshaw, A. Chemistry of the Elements; Pergamon
Press: New York, 1984, 1443.

354

Jl111111111llllllll!1111111llllllllllllWHI