-»m«. MM“ '- ..-,g.. ~.’ku \H”: ‘ ml" 313:}. fi ...: .v.;;.;:”_;mr - (-1 l ' I-‘ I'- .. v. ._ 7" r”. ~ IHW.’ ‘ n. :n .1..- . :9“, .. . Illilllllll'il lllllllllllllllll lllllllll 3 1293 01049 0963 This is to certify that the dissertation entitled LON-TEMPERATURE SYNTHESIS OF NEW TERNARY CHALCOGENIDE COMPOUNDS OF Cu, Au, AND Hg USING ALKALI METAL POLYCHALCOGENIDE FLUXES presented by Younbong Park has been accepted towards fulfillment of the requirements for PhD degree in [men] j stny Wu #2; (11/) M ajor professor Q/N/‘fL I Date MSU is an Affirmative Action I'Equal Opportunity Institution 0-12771 LiSRARY { l Michigan State :- University PLACE lN RETURN BOX to remove We checkout from your tecord. TO AVOID FINES tetum on or before date due. DATE DUE DATE DUE DATE DUE MSU I. An Affirmative ActiorVEquai Opportunity Institution cMMms-oi LOW-TEMPERATURE SYNTHESIS OF NEW TERNARY CHALCOGENIDE COMPOUNDS OF Cu, Au, AND Hg USING ALKALI METAL POLYCHALCOGENIDE FLUXES BY Younbong Park A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1 992 ABSTRACT LOW-TEMPERATURE SYNTHESIS OF NEW TERNARY CHALCOGENIDE COMPOUNDS OF Cu, Au, AND Hg USING ALKALI METAL POLYCHALCOGENIDE FLUXES BY Younbong Park In last two decades great efforts have been exerted to find new materials with interesting optical, electrical, and catalytic properties. Metal chalcogenides have been studied extensive because of their interesting physical properties and rich structural chemistry, among the potential materials. Prior to this work, most known metal chalcogenides had been synthesized at high temperature (T>500 oC). Intermediate temperature synthesis in solid state chemistry was seldom pursued because of the extremely slow diffusion rates between reactants. This intermediate temperature regime could be an new synthesis condition if one looks for new materials with unusual structural features and properties. Metastable or kinetically stable compounds can be stabilized in this intermediate temperature regime, in contrast to the thermodynamically stable high temperature compounds. Molten salts, especially alkali metal polychalcogenide fluxes, can provide a route for exploring new chalcogenide materials at intermediate temperatures. These fluxes are very reactive and melt as low as 145 OC (mp of K284). Using these fluxes as reaction media, we have encountered many novel chalcogenide compounds with unusual structures and interesting electrical properties (semiconductors to metallic conductors). Low-dimensional polychalcogenide compounds of a-ACuQ4(A=K, Cs; Q=S, Se), B-KCuS4, KAuQs (Q=S, Se), K3AuSe13, Na3AuSe3, and CsAuSea exhibit the beautiful structural diversity and bonding flexibility of the polychalcogenide ligands. In addition, many novel chalcogenide compounds of Cu, Hg, and Au with low- dimensional structures. The preparation of novel mixed-valence Cu compounds, K20U5Te5, Csch3Te1o, Na30U4Se4, K30U384Te2, and KCU4SZTe, which show interesting metallic properties, especially underscores the enormous potential of the molten salt method for the synthesis of new chalcogenide materials with interesting physical properties. The materials prepared in this study can be classified as a new class of chalcogenide compounds due to their unique structures. In this dissertation the synthesis, characterization with emphasis on structures, charge transport properties, and magnetic susceptibilities of the materials will be illustrated. To my lovely wife, Jiyeon, and son, Sangjune iv ACKNOWLEDGEMENTS None of the work described in this dissertation would have been possible without the dedication, patient guidance and support of my research adviser, Professor Mercouri G. Kanatzidis. I would also thank the other members of my committee: Drs. Babcock, Jackson, and especially Eick for his helpful comments as a second reader. I also thank Professor Kannewurf for charge transport property measurements, Professor Cowen for his helpful comments on magnetic susceptibility measurements, and Dr. Ward for helpful discussions on X-ray single crystal study. I would like to extend my deepest gratitude and heart-felt thanks to all the Kanatzidis group members, who made graduate school an enjoyable learning expenence. Most importantly I am grateful to my wife, .Jiyeon, and son, Sangjune for all the love and understanding they have given me over the years. Their patience and encouragements were essential to the completion of this dissertation. I like to extend my love and thanks for being so supportive to my parents and parent-in-Iaws who have been behind me always. Financial support given by BASF summer fellowship (1991), Center for Fundamental Material Research, the Department of Chemistry at Michigan State University, and National Science Foundation is greatly acknowledged and appreciated. vi TABLE OF CONTENTS Page List of Tables .............................................................................................. xv List of Figures .............................................................................................. xxiii I. Introduction .............................................................................................. 1 1. Discrete Molecular Metal Chaloogenide Chemistry ..................... 1 2. High Temperature Metal Chaloogenide Chemistry ...................... 6 3. Molten Salt Synthetic Methods ......................................................... 12 4. New Materials from Polychalcogenide Flux at Intermediate Temperatures ........................................................... 1 7 ll. Synthesis and Characterization of Low-Dimensional (Poly)chalcogenide Compounds in the A/Cu/Q System (A=Na, K, Cs; Q=S, Se) ............................................................................. 19 1. Introduction ........................................................................................... 19 2. Experimental Section ........................................................................ 20 2.1. Reagents ...................................................................................... 20 2.2. Physical Measurements ............................................................ 21 2.3. Synthesis ..................................................................................... 22 Potassium sulfide, K28 ............................................................... 22 Cesium selenide, Cszse ............................................................ 23 a-Potassium (1 ,2-p2-tetrasulfido)cuprate(I), a-KCuS4 (I) ..... 23 vii Page p-Potassium (1,4-p2-tetrasulfido)cuprate(I), p-KCuS4 (II) .................................................................................... 24 a-Potassium (1 ,2-p2-tetraselenido)cuprate(I), a—KCuSe4 (Ill) ................................................................................ 24 a-Cesium (1 ,2-p2-tetraselenido)cuprate(l), a-CSCUSe4 (IV) ............................................................................. 24 Trisodium tetra(p3-selenido)tetracuprate(I,II), NaaCu4$e4 (V) .............................................................................. 25 Sodium bis(p4-selenido)dicuprate(I)—copper oxide N81,90U2 8820120 (VI) ............................................................. 25 2.4. X-ray Crystallographic Studies ................................................ 26 Results and Discussion ..................................................................... 40 3.1. Synthesis and Spectroscopy ................................................... 40 3.2. Description of Structures ........................................................... 47 3.2.1. Structure of a-KCuS4(l), a-KCuSe4(III), and a-CsCuSe4(IV) .......................................................... 47 3.2.2. Structure of p-KCuS4 (ll) .................................................. 54 3.2.3. Structure of NagCU48e4 (V) ............................................ 58 3.2.4. Structure of Na1,90que20uzO (VI) ............................ 63 3.3. Close Structural Relationships between a- and fi- [Cu(S4)]n"' Chains ..................................................................... 68 3.4. Charge Transport Properties of a-KCuSe4 (III) and Na1,90uzse20uz0 (VI) ........................................................... 71 viii Page III. Low-Dimensional Compounds Incorporating (Poly)chalcogenide Ligands in the AlAu/Q (A=Na, K, Cs; Q=S. Se) Systems ............................................................. 76 1. Introduction ........................................................................................... 76 2. Experimental Section ......................................................................... 77 2.1. Reagents ..................................................................................... 77 2.2. Physical Measurements ............................................................ 77 2.3. Synthesis .................................................................................... 78 Potassium (1 ,5-p2—pentasulfido)aurate(I), KAuSs (I) ......................................................................................... 79 Potassium (1 ,5-p2-pentaselenido)aurate(I), KAuSes (II) ..................................................................................... 79 Cesium (1 ,3-p2-triselenido)aurate(I), CsAuSe. (III) .................................................................................. 79 Tripotassium bis(pentaselenido) (1,3-p2-triselenido)aurite(llI), K3AuSe13 (IV) ......................... 80 Potassium bis(u2-selenido)aurite(lll), KAuSez (V) ................ 80 Trisodium bis(triselenido)(pg-diselenido) aurite(|ll), NaaAuSea (VI) ............................................................. 81 Potassium bis(p2-selenido)aurite(lll), NaAuSez (VII) ................................................................................ 81 2.4. X-ray Crystallographic Studies ................................................ 82 3. Results and Discussion ..................................................................... 101 3.1. Synthesis and Spectroscopy ................................................... 101 3.2. Description of Structures ........................................................... 107 3.2.1. Structure of KAuSs (I) and KAuSe5 (II) .......................... 107 3.2.2. Structure of CsAuSes (III) ................................................. 1 12 Page 3.2.3. Structure of KgAuSe13 (IV) .............................................. 1 15 3.2.4. Structure of KAuSez (V) ................................................... 120 3.2.5. Structure of NagAuSea (VI) ............................................. 122 3.2.6. Structure of NaAuSeg (VII) .............................................. 1 26 3.3. Structures ..................................................................................... 130 IV. Molten Salt Synthesis of Low-Dimensional Ternary Chalcogenides. Novel Structure Types in the AIHg/Q System (A=K, Cs; Q=S, Se) ...................................................................... 136 1. Introduction ........................................................................................... 136 2. Experimental Section ........................................................................ 137 2.1. Reagents ..................................................................................... 137 2.2. Physical Measurements ............................................................ 137 2.3. Synthesis ..................................................................................... 138 Dipotassium tetra(p2-suIfido)trimercurate(ll), K2H9384 (I) ..................................................................................... 138 Dipotassium tetrawz-selenido)trimercurate(ll), Kai-I93Se4(ll) ................................................................................. 139 Dicesium tetra(p2-selenido)trimercurate(II), Gael-@8940") .............................................................................. 139 Dipotassium heptamz-sulfidomexamercurate(II), Kai-9687 (IV) .................................................................................. 140 Dicesium hepta(p2-selenido)hexamercurate(lI), CsQI-IQSSe; (V) .............................................................................. 140 2.4. X-ray Crystallographic Studies ................................................ 141 3. Resultsand Discussion ...................................................................... 153 3.1. Synthesis ..................................................................................... 153 Page 3.2. Description of Structures ........................................................... 155 3.2.1. Structure of K2H9384(l), K2H938e4(ll), amchdfihSSNM) ........................................................... 155 3.2.2. Structures of K2H95$7 (IV) and CSzngse7 (V) ......... 162 33.8humwes ..................................................................................... 170 V. Synthesis and Characterization of Novel Layered Compounds of KZCusTe5 and NaCuTe ........................................................................ 172 1. Introduction ........................................................................................... 172 2. Experimental Section ........................................................................ 173 2J.Fk£gans ...................................................................................... 173 2.2. Physical Measurements ............................................................ 173 2.3. Synthesis ..................................................................................... 175 Dipotassium (us-ditelIuro)tris(,u4-telluro) mamxmmabUMLIQCmflqfiD ................................................. 175 Sodium (p4-telluro)cuprate(l), NaCuTe (ll) .............................. 175 2.4. X-ray Crystallographic Studies ................................................ 176 Results and Discussion ...................................................................... 182 3.1. Synthesis .................................................................................... 182 3.2. Description of Structures ........................................................... 183 3.2.1. Structure of K20U5Te5(l) ................................................. 183 3.2.2. Structure of NaCuTe (II) ................................................... 190 3.3. Structural Relationships of the [CusTeslnzn' Layers in (I) to Those of [CuTe]n"' in (II) and CuTe ........................... 192 3.4. Charge Transport Properties of K2CU5T65 (I) ...................... 194 3.5. Magnetic Susceptibility of chusTes (I) ................................. 197 Page VI. Synthesis and Characterization of K4CU3T611 and 033CU3Te1o: Novel Solid State Chaloogenide Compounds with a Dodecahedral Cluster as a Building Block ............................................ 200 1. Introduction ........................................................................................... 200 2. Experimental Section ......................................................................... 201 2.1. Reagents ...................................................................................... 201 2.2. Physical Measurements ............................................................ 201 2.3. Synthesis ..................................................................................... 203 Tetrapotassium bis(p4citelluro)tris(p3-ditelluro) ([14-telluro) octacuprate(l), K40U3Te11 (I) ................................ 203 Tricesium bis(p4-ditelluro)bis(p8-ditelluro) bis(p4-telluro) octacuprate(l,ll), 0530uaTe1o(ll) ..................... 203 2.4. X-ray Crystallographic Studies ................................................ 204 3. Results and Discussion ...................................................................... 212 3.1. Synthesis ..................................................................................... 212 3.2. Description of Stmctures ........................................................... 214 3.2.1. Structure of K40ueTe11 (I) ............................................... 214 3.2.2. Structure of 053CuaTe10 (ll) ............................................ 223 3.3. Comparison of Cua(Te2)5 Cluster Size in K40U3Te11 (I) and CsoCueTew (ll) ........................................ 228 3.4. Charge Transport Properties of K40U3Te11 (I) and 03301.13Te10 (II) .................................................................. 231 3.5. Magnetic Susceptibility of CSQCU8T910 (II) ........................... 236 Page VII. Synthesis and Characterization of Mixed Chaloogenide Compounds of KCU482Te, K30U3$4Te2, and Cu17,5Te3826 ........... 238 1. Introduction ........................................................................................... 238 2. Experimental Section ......................................................................... 239 2.1. Reagents ..................................................................................... 239 2.2. Physical Measurements ............................................................ 239 2.3. Synthesis ..................................................................................... 243 Potassium bis(p4-sulfido)(p3-telluro) tetracuprate(l,ll), KCu4SZTe (I) ................................................... 243 Tripotassium tetra(u4-suIfido)bis(p5-telluro) octacuprate(l,ll) recitesnez (Il) ............................................... 243 Copper tellurium sulfide, Cu17,6Te3825(lll) ........................... 244 2.4. X-ray Crystallographic Studies ................................................ 244 3. Results and Discussion ...................................................................... 253 3.1. Synthesis ..................................................................................... 253 3.2. Description of Structures ........................................................... 255 3.2.1. Structure of KCU4SzTe (I) ................................................ 255 3.2.2. Structure of K30u884Te2(ll) ........................................... 261 3.2.3. Structure of Cu17_6Te3326(III) ....................................... 268 3.3. Charge Transport Properties of KCmSzTe (I) ..................... 273 3.4. Magnetic Susceptibility of KCU482Te (I), K3CUeS4T62 (II), and CU17,5T83826 (III) ............................... 276 Page VIII. AuCuSe4, a Novel Mixed Metal Chaloogenide Compound Incorporating Sea? ligands ...................................................................... 283 1. Introduction ........................................................................................... 283 2. Experimental Section ........................................................................ 234 2.1. Reagents ...................................................................................... 284 2.2. Physical Measurements ............................................................ 284 2.3. Synthesis .................................................................................... 285 Copper gold tetraselenide, AuCuSe4(l) .................................. 285 2.4. X-ray Crystallographic Studies ................................................ 286 3. Results and Discussion ...................................................................... 291 3.1. Synthesis and Spectroscopy ................................................... 291 3.2. Description of Structure ............................................................. 294 IX. Conclusion ................................................................................................... 302 List of References .................................................................................................. 304 xiv 10. 11. LIST OF TABLES Page Types of Metal Polychalcogenide Complex Containing Q22' ............. 3 Types of Metal Polychalcogenide Complex Containing 053' (X 23) ...................................... 5 Melting Points (00) of Some Known Alkali Metal Polychalcogenides ...................................................................................... 16 Calculated and Observed Powder Diffraction Pattern for a-KCUS4. ................................................................................................. 27 Calculated and Observed Powder Diffraction Pattern for B-KCuS4 .................................................................................................. 28 Calculated and Observed Powder Diffraction Pattern for a-KCuSe4 ............................................................................................... 29 Calculated and Observed Powder Diffraction Pattern for or-CSCUSe4. ............................................................................................. 30 Calculated and Observed Powder Diffraction Pattern for NmCmSea ............................................................................................ 31 Calculated and Observed Powder Diffraction Pattern for Na1,QCque20uzO ............................................................................... 32 Summary of Crystallographic Data for a-KCuS4, p-KCuS4, a-KCuSe4, a-CsCuSe4, Na3CU4Se4, and N313CU2382'CU20 ........ 35 Fractional Atomic Coordinates and Beq Values for a-KCuS4 A with Their Estimated Standard Deviations in Parentheses .................. 37 XV 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Page Fractional Atomic Coordinates and ng Values for SKCuS4 with Their Estimated Standard Deviations in Parentheses .................. 37 Fractional Atomic Coordinates and ng Values for a-KCuSe4 with Their Estimated Standard Deviations in Parentheses .................. 38 Fractional Atomic Coordinates and ng Values for a-CsCuSe4 with Their Estimated Standard Deviations in Parentheses .................. 38 Fractional Atomic Coordinates and ng Values for NamCU48e4 with Their Estimated Standard Deviations in Parentheses .................. 39 Fractional Atomic Coordinates and ng Values for Na1,90uzse2-CU20 with Their Estimated Standard Deviations in Parentheses ......................................................................... 39 Frequencies (cm'l) of Spectral Absorption of (I), (II), (III), and (IV) Due to 0-0 and M-0 Stretching Vibrations .............................. 46 Selected Bond Distances (A) and Angles (Deg) in a-KCUS4 with Standard Deviations in Parentheses ............................................... 51 Selected Bond Distances (A) and Angles (Deg) in a-KCuSe4 with Standard Deviations in Parentheses ............................................... 52 Selected Bond Distances (A) and Angles (Deg) in a-CsCuSe4 with Standard Deviations in Parentheses ............................................... 53 Selected Bond Distances (A) and Angles (Deg) in 8-KCuS4 with Standard Deviations in Parentheses ............................................... 57 Selected Bond Distances (A) and Angles (Deg) in Na3CU4Se4 with Standard Deviations in Parentheses ............................................... 62 Selected Bond Distances (A) and Angles (Deg) in Na1,9Cuzse2-Cuzo with Standard Deviations in Parentheses .......... 66 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. Page Calculated and Observed Powder Diffraction Pattern for KAUSs ....................................................................................................... 83 Calculated and Observed Powder Diffraction Pattern for KAu835 ................................................................................................... 84 Calculated and Observed Powder Diffraction Pattern forCsAuSeg .................................................................................................. 85 Calculated and Observed Powder Diffraction Pattern for K3AUSS13. ............................................................................................... 86 Calculated and Observed Powder Diffraction Pattern for KAuSez .................................................................................................... 88 Calculated and Observed Powder Diffraction Pattern for NagAuSeg ............................................................................................... 89 Calculated and Observed Powder Diffraction Pattern for NaAuSez .................................................................................................. 91 Summary of Crystallographic Data for KAuSs, KAuSes, CsAuSea IQAuSe13, KAuSez, NagAuSeg, and NaAuSeg ................. 93 Fractional Atomic Coordinates and 89q Values for KAuSs with Their Estimated Standard Deviations in Parentheses .................. 96 Fractional Atomic Coordinates and ng Values for KAuSes with Their Estimated Standard Deviations in Parentheses .................. 96 Fractional Atomic Coordinates and ng Values for CsAuSea with Their Estimated Standard Deviations in Parentheses .................. 97 Fractional Atomic Coordinates and Beq Values for KgAuSe13 with Their Estimated Standard Deviations in Parentheses .................. 98 xvii 36 37. 38 39. 40. 41. 42. 45. 46. 47. Page Fractional Atomic Coordinates and ng Values for KAuSez with Their Estimated Standard Deviations in Parentheses .................. 99 Fractional Atomic Coordinates and ng Values for Na3AuSe8 with Their Estimated Standard Deviations in Parentheses ................. 99 Fractional Atomic Coordinates and ng Values for NaAuSez with Their Estimated Standard Deviations in Parentheses .................. 100 Frequencies (cm-1) of Spectral Absorption of (I), (II), (III), (IV), (V), (VI), and (VII) Due to 0-0 and M-0 Stretching Vibrations ............. 103 Selected Bond Distances (A) and Angles (Deg) in KAuSs and KAuSes with Standard Deviations in Parentheses ....................... 111 Selected Bond Distances (A) and Angles (Deg) in CsAuSea with Standard Deviations in Parentheses ............................................... 114 Selected Bond Distances (A) and Angles (Deg) in K3AuSe13 with Standard Deviations in Parentheses ............................................... 119 Selected Bond Distances (A) and Angles (Deg) in KAuSez with Standard Deviations in Parentheses ............................................... 120 Selected Bond Distances (A) and Angles (Deg) in Na3AuSe3 with Standard Deviations in Parentheses ............................................... 124 Selected Bond Distances (A) and Angles (Deg) in NaAuSez with Standard Deviations in Parentheses ............................................... 128 Calculated and Observed Powder Diffraction Pattern for Kzl-Igas4. ................................................................................................. 142 Calculated and Observed Powder Diffraction Pattern for Kai-bases ............................................................................................... 143 48. 49. 51. 52. 55. 57. 58. 59. Page Calculated and Observed Powder Diffraction Pattern for CszHg3$e4 ............................................................................................. 144 Calculated and Observed Powder Diffraction Pattern for K2H99$7 .................................................................................................. 145 Calculated and Observed Powder Diffraction Pattern for Cszl-IgeSey .............................................................................................. 1 46 Summary of Crystallographic Data for K2H9334, K2H938e4, CszHgasea, K2H9637, and CSzngSe7 ................................................. 148 Fractional Atomic Coordinates and ng Values for K2H9384 with Their Estimated Standard Deviations in Parentheses .................. 150 Fractional Atomic Coordinates and ng Values for K2H93Se4 with Their Estimated Standard Deviations in Parentheses .................. 150 Fractional Atomic Coordinates and ng Values for CszH93$e4 with Their Estimated Standard Deviations in Parentheses .................. 151 Fractional Atomic Coordinates and Beq Values for K2H9587 with Their Estimated Standard Deviations in Parentheses .................. 151 Fractional Atomic Coordinates and Beq Values for CSzHgsSe'] with Their Estimated Standard Deviations in Parentheses .................. 152 Selected Bond Distances (A) and Angles (Deg) in K2H93$4 with Standard Deviations in Parentheses ............................................... 159 Selected Bond Distances (A) and Angles (Deg) in K2Hgase4 with Standard Deviations in Parentheses ............................................... 160 Selected Bond Distances (A) and Angles (Deg) in CszHgase4 with Standard Deviations in Parentheses ............................................... 161 60. 61. 62. 65. 66. 67. 68. 69. 70. 71. 72. Page Selected Bond Distances (A) and Angles (Deg) in K2H9537 with Standard Deviations in Parentheses ............................................... 168 Selected Bond Distances (A) and Angles (Deg) in CszngSe7 with Standard Deviations in Parentheses ............................................... 169 Calculated and Observed Powder Diffraction Pattern for K2015Te5 ................................................................................................ 177 Calculated and Observed Powder Diffraction Pattern for NaCuTe .................................................................................................... 178 Summary of Crystallographic Data for KZCusTes and NaCuTe ........ 180 Fractional Atomic Coordinates and ng Values for K2CU5Te5 with Their Estimated Standard Deviations in Parentheses .................. 181 Fractional Atomic Coordinates and &q Values for NaCuTe with Their Estimated Standard Deviations in Parentheses .................. 181 Selected Bond Distances (A) and Angles (Deg) in K2CU5T65 with Standard Deviations in Parentheses ............................................... 189 Selected Bond Distances (A) and Angles (Deg) in NaCuTe with Standard Deviations in Parentheses ............................................... 190 Calculated and Observed Powder Diffraction Pattern for I<40tlgTe11 ............................................................................................. 205 Calculated and Observed Powder Diffraction Pattern for 083Cu3Te10 .......................................................................................... 207 Summary of Crystallographic Data for K40U3Te11 and CmCtlgTeto ................................................................................................ 209 Fractional Atomic Coordinates and ng Values for K40U3Te11 with Their Estimated Standard Deviations in Parentheses ................. 210 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. Page Fractional Atomic Coordinates and ng Values for C53CueTeto with Their Estimated Standard Deviations in Parentheses ................. 211 Selected Bond Distances (A) and Angles (Deg) in K4CU3T911 with Standard Deviations in Parentheses .............................................. 220 Selected Bond Distances (A) and Angles (Deg) in CS30u8Te10 with Standard Deviations in Parentheses .............................................. 227 Selected Metric Data for the KCua(Te2)6 Cluster (A) and the CSCI.Ia(Tez)6 Cluster (B) ...................................................................... 229 Calculated and Observed Powder Diffraction Pattern for KCU432Te .............................................................................................. 245 Calculated and Observed Powder Diffraction Pattern for K30ueS4Te2 .......................................................................................... 246 Calculated and Observed Powder Diffraction Pattern for Cling-[89325 ........................................................................................ 247 Summary of Crystallographic Data for KCU482Te, K3CUgS4Te2, and Cu17,6Te5826 ............................................................. 250 Fractional Atomic Coordinates and ng Values for KCU482Te with Their Estimated Standard Deviations in Parentheses ................. 251 Fractional Atomic Coordinates and ng Values for IQCU3$4Te2 with Their Estimated Standard Deviations in Parentheses ................. 251 Fractional Atomic Coordinates and Beq Values for Cur/,GTegst with Their Estimated Standard Deviations in Parentheses ................ 252 Selected Bond Distances (A) and Angles (Deg) in KCU482Te with Standard Deviations in Parentheses .............................................. 260 85. 86. 87. 89. 90. Page Selected Bond Distances (A) and Angles (Deg) in K3Cu8S4Te2 with Standard Deviations in Parentheses .............................................. 266 Selected Bond Distances (A) and Angles (Deg) in Cu17_6Te8$26 with Standard Deviations in Parentheses .............................................. 272 Calculated and Observed Powder Diffraction Pattern forAuCuSea ............................................................................................... 287 Summary of Crystallographic Data for AuCuSea, ................................ 289 Fractional Atomic Coordinates and ng Values for AuCuSe4 with Their Estimated Standard Deviations in Parentheses ................. 290 Selected Bond Distances (A) and Angles (Deg) in AuCuSe4 with Standard Deviations in Parentheses .............................................. 300 LIST OF FIGURES Page 1. ORTEP representation of the unit cell of (A) Zinc-blende (ZnS) and (B) Wurtzite (ZnS). Open circles are S atoms and octant-shaded ellipsoids are Zn atoms. ................................................. 8 2. ORTEP representation of the unit cell of (A) TiS (NlAs-type) and (B) TiSz (Cdlz-type). Open circles are S atoms and octant-shaded ellipsoids are Ti atoms. .................................................. 9 3. ORTEP representation of the unit cell of (A) Pyrite (FeSz) and (B) Marcasite (FeSz). Open circles are S atoms and octant-shaded ellipsoids are Fe atoms. ................................................ 1 1 Phase diagrams of (A) NagSIS47 and (B) KQS/S48 systems. ............ 14 Far-IR spectra of (A) a-KCuS4. (B) p-KCuS4, (C) a-KCuSe4, and (IV) a-CsCuSe4. ................................................................................. 44 6. Stereoview of the unit cell of a-ACuQ4 (A=K, Cs; O=S, Se). The alkali atoms have been omitted for clarity. .................................... 48 7. ORTEP representation of the one-dimensional infinite a-[CuQ4 "' chain with labeling scheme. .............................................. 49 8. Stereoview of the unit cell of B-KCuS4. The alkali atoms have been omitted for clarity. .................................... 55 9. ORTEP representation of the one-dimensional infinite p—[CuSdn ' chain with labeling scheme. ............................................... 56 10. ORTEP representation of the unit cell of N330u.tSe4. ........................ 59 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21 . Page Two views of the one—dimensional structure of [CU4Se4]n3"' with labeling scheme. ................................................................................ 60 ORTEP representation of the unit cell of Na1,9CU2362CU20 with labeling scheme. ................................................................................ 64 ORTEP representation of (A) [CuSe]n ' layer (B) CuzO layer viewed down [001] direction. Octant-shaded ellipsoids are Cu atoms and Open circles are Se and O atoms. ......................... 65 Schematic representation of the structural transformation of (A) the ot-[CuS4lnn' chain to (D) the B-[CuS4lnn’ chain via (8) the hypothetical [CuS41nn' intermediate which is similar in structure to (C) the known [AgSe4]n"° chain. ........................................................ 69 Four probe electrical conductivity (Scm'l) data as a function of temperature for a single crystal of a-KCuSe4. ...................................... 72 Four probe resistivity (yo-cm) data as a function of temperature for a single crystal of Na1,gCuzse20uzo. ............................................ 73 Thermoelectric power (pV/K) data as a function of temperature for a single crystal of Na1,gCque2Cuzo. ............................................ 74 Far-IR spectra of (A) KAuSs, (B) CsAuSea, and (C)KAuSe5 . ........... 104 Far-IR spectra of (A) KgAuSets and (B) NagAuSeg ............................ 105 Far-IR spectra of (A) KAuSez and (B) NaAuSez. ................................. 106 ORTEP representation of the unit cell of KAUCE (Q=S, Se). ............. 108 One-dimensional structure of [Au05]n"' chains, showing the mode of dimerization. ................................................................................ 1 10 xxiv 23. 24. 25. 26. 27. 28. 30. 31. 32. 33. 34. 35. 36. Page ORTEP representation of two axial views of CsAuSe3: (A) viewed down [100] direction, (B) viewed down [01 1] direction. ............................................................................................ 1 13 Stereoview of the unit cell of KgAuSeta. ............................................... 1 16 ORTEP representation of two views of the one-dimensional [AuSei 3]n3"' chain. .................................................................................... 1 1 7 Coordination environments of K atoms with labeling scheme. ......... 1 18 ORTEP representation of the unit cell of KAuSez. ............................... 121 ORTEP representation of two views of the one-dimensional [AuSegln3n‘ chain. ...................................................................................... 1 23 Coordination environments of Na atoms with labeling scheme. ....... 125 ORTEP representation of the layered structure of [AuSe2]n"', looking perpendicular to the layer. ......................................................... 127 ORTEP representation of the unit cell of NaAuSeg. ............................ 129 Schematic representation of the counter cation size effect on structures of [AuSezlnn'z (A) layer structure of [AuSezlnn' in (VII) vs (B) chain structure of [AuSezinn' in (V). ............................................. 134 Two axial views of the unit cell of A2H9304 (A=K, Cs; Q=S, Se): (A) b—axial view and (B) projection view along (001) plane. .............. 156 Two views of the one-dimensional structure of [HggQ4]n2n- with labeling scheme. ................................................................................ 157 Stereoview of the unit cell of K2H9587. ................................................. 163 Stereoview of the unit cell of C82H95887. ............................................ 164 37. 38. 39. 40. 41. 42. 45. 46. ORTEP representation of (A) the unit cell of K2H9667 and (B) the unit cell of CSzngSe7. The origin of the unit cell of CSzHgaSe7 has been moved to (1/2, 0, 0) for clear comparison. ORTEP representation and labeling scheme of the unit cell of K20u5T35 ........................................................................... Structural relationship of the [CusTeslnzn' layer to those of CuTe (above) and [CuTe1nn' (below) as a function of oxidation state. .................................................................................. Two axial views of the layered structure of CuTe: (A) a-axial view and (B) baxlal view. ................................................ ORTEP representation of the layered structure of NaCuTe. ......... ORTEP representation of (A) the [CuTelnn' layer and (B) the unit cell of KCuTe. .......................................................................... Four probe resistivity (pa—cm) data as a function of temperature for a single crystal of K20U5Te5. .................................. Variable-temperature thermoelectric power (pV/K) data for a single crystal of K20u5Te§ .............................................................. Variable-temperature magnetic susceptibility (emu/moi) data for polycrystalline chusTes. Insert: Xm" (moi/emu) vs T. ................. ORTEP representation of the unit cell of K40U3Te11 viewed down [010] direction. ............................................................. xxvi Page ..... 165 ..... 184 ..... 185 ..... 186 ..... 191 ..... 193 195 ..... 196 ..... 198 ..... 215 47. 48. 49. 51. 52. 55. ORTEP representation and labeling scheme of (A) the [KCuaTe12] dodecahedral cluster, (8) two edge-shared [KCuaTetz] units with capping Te?“ ions, (C) a one-dimensional column of edge-shared double clusters. and (D) the empty CugTeg cluster. ....................................................... ORTEP representation of (A) sideview and (8) top view of one Cull“ e layer. The Cua(Te2)6 and CuaTeg clusters are shaded for emphasis. ....................................................... ORTEP representation of the unit cell of 033CuaTe10. .................... ORTEP representation of (A) sideview and (B) top view of one [CuaTetolni’tn' layer. The Cu8(Te2)6 and CuaTeg clusters are shaded for emphasis. ....................................................... Four probe electrical conductivity (Scm'1) data as a function of temperature for a single crystal of K4Cu8Te11. ........... Four probe resistivity (pa-cm) data as a function of temperature for a Single crystal of CSgCtlgTem ....................................................... Variable-temperature thermoelectric power (pV/K) data for a single crystal of ngCugTem ............................................................ Variable-temperature magnetic susceptibility (emu/mo!) data for polycrystalline CSQCUgTem. Insert: 30““ (moi/emu) vs T. ............... Magnetization (emu) data for polycrystalline Cu17,6Tegszs as a function of applied magnetic field (H). ........................................ xxvii Page 216 219 224 225 .. 232 .. 233 235 .. 237 241 56. 57. 58. 59. 61. 62. 65. 66. 67. Page ORTEP representation of (A) the unit cell and (B) the layered framework of KCU482Te. The origin of the unit cell has been moved to (0,0,112) in (A) for clear view of a building unit. .................. 256 ORTEP representation of the [CU482Te]n"' layer. Black circles are Te, open circles are S, and crossed circles are Cu atoms. ......... 257 ORTEP representation of the unit cell of IQCueSltTeg, ,,,,,,,,,,,,,,,,,,,,,,, 262 ORTEP representation and labeling scheme of the [CuaS4Te2]n3"' layer. ................................................................................ 263 ORTEP representation and labeling scheme of the unit cell of CunfiTeeSgs ............................................................................................ 269 Coordination environments of (A) Cu(1) atom, (B) Cu(2) atom, (C) Te atom, (D) S(1) atom, and (E) 8(2) atom in Cu17,6T33$26. ...................................................... 271 Four probe resistivity (pO-cm) data as a function of temperature for a single crystal of KCu4$2Te ............................................................ 274 Variable-temperature thermoelectric power (pV/K) data for a single crystal of KCu462Te. .................................................................. 275 Variable-temperature magnetic susceptibility (emu/moi) data for polycrystalline KCU482Te. Insertzxm'1 (moi/emu) vs T. ..................... 277 Variable-temperature magnetic susceptibility (emu/moi) data for polycrystalline K30u884Te2. Insert: xm" (moi/emu) vs T. ................. 278 Variable-temperature magnetic susceptibility (emu/moi) data for polycrystalline Cu17_6Te3625 ................................................................. 280 Temperature dependence of calculated magnetic moments (BM) of CU17,6Te3$26 .............................................................................. 281 xxviii 69. 70. 71 . Page Far-IR spectrum of AuCuSea, .................................................................. 293 ORTEP representation of the unit cell of AuCuSe4. Crossed circles are Cu atoms, black circles are Au, and open circles are Se atoms. ............................................................... 295 Two views of the [Au(Se3)(Se)]n"' chains with labeling scheme. ..... 296 ORTEP representation and labeling scheme of one layer fragment of AuCuSea ..................................................................... 297 CHAPTER 1 Introduction For the last two decades the synthetic chemistry of new metal chalcogenide compounds has been developed considerably due to their interesting electrical, optical, and catalytic properties as well as unusual structural features. Most metal chalcogenide compounds known to date have been synthesized either at ambient temperature or at temperatures higher than 500 001.2. However, there is increased interest in new and unusual synthetic conditions which may help to stabilize new chalcogenide materials with interesting physical properties and unusual structure that would not be possible by conventional techniques. One possible synthesis approach we have undertaken is the use of molten salts at intermediate temperature regime (150 00 < T < 500 oC). 1. Discrete Molecular Metal Chaloogenide Chemistry Molecular metal chalcogenide chemistry, especially metal sulfide chemistry, has been developed extensively due to (a) potential application as hydrodesulfurization catalysts3v4, (b) the importance of some MIS complexes in modeling bioinorganic systems5, and (c) the structural diversity associated with polysulfide (sz') ligands. Recently there also have been increased interests in the heavier chalcogenide (Se, Te) compounds.6.7 This has been stimulated 2 mainly by their distinct structural chemistry and potential uses as low temperature precursor compounds to either new or old, but useful electronic solid state materials5u3. Molecular metal chalcogenide compounds generally have been synthesized by reacting metal salts with alkali metal polychalcogenides (AzQx) in polar non-aqueous solvents at ambient temperature as shown in eq 1.6- 7(a) MCI“ + nNaQQx ----> [M(oq)y]m- + NaCl eq 1. The anionic M/Q complexes are crystallized from solution with large organic cations such as R4N+ and R4P+. Less common method such as the reaction of low valent metal carbonyls with a polychalcogenide source also has been used.7(blv9 Molecular metal chalcogenide compounds exhibit structural diversity featuring various polychalcogenide species, from monochalcogenides (02'; in [NazFe1383018'-10) to nanochalcogenides (092'; in [Au(Sg)]1'-11). The usual binding modes of polychalcogenides are numerous and are given in Tables 1 and 2. They can assume bridging or chelating modes or bind to multiple metals simultaneously. The internal O atoms of the sz' ligands are considered to have nearly zero formal charges and will be less likely to participate in bonding interactions with other metals than the negatively charged terminal 0 atoms. However, it has been found that not only terminal atoms but also internal atoms are available for bonding to metal ions. The ability of chalcogens to bind to multiple metal centers and to catenate, gives rise to the enormous structural diversity and bonding characteristics of the polychalcogenide compounds. 3 Table 1. Types of Metal Polychalcogenide Complexes Containing 022' Type Example Reference Q M< I (Ph4P)2[W23¢4(S°2)(S¢4)] 12 Q M Q/ M<| (nS-C5H5)Fc(c0)2(8e2)Cr(co)2(n5-CSH5) 13 Q Q M< I > M (n5'C5H5)2CF2(C0)4S°2 14 Q M \Q Q M/ M Q/ I [Ins-C5H4Me)2Tilz(Sez)2 ' 16 Table 1. (cont'd) Type Example Reference Q I [K-2.2.2-crypt12[M04(T62)5(T<>3)2(en)4l 17 Q M/ \M M M \Q/ l [NBU4I4IH84T312] 18 M M M Q/ l [{(ns-C5H5)(C0)2Fe}3(3t‘2)l[BF4] 19 M/Q\M M.— l M [(C0)5W(ll-T62)][W(C0)s]2 20 5 Table 2. Types of Metal Polychalcogenide Complexes Containing sz‘ (x 2 3) Type 0,? Example Reference Qx x=3 (Ph4P)2[Auz(Se5)(Se¢)] 21 K \ x=4 (Ph4P)2[Au2(Se7)(Se4)2] 21 M M x=5 (Ph4P)2 [102(864) 4(S65)] 22 (QMW x: (Ph4P)2[WZSe4(Se3),J 12 Q Q x=4 [(Ph3 P)2N] 2[AUZS%(SC4)2] 23 \M/ x=5 (Ph4P)2[FezSe2(Ses)2] 24 Qx-Z ( W F“ (P114?)2[H82(S€4)3] 25 Q\ /Q\M M on2 / \Q X=4 (PrthiA84(Sea)3] 26 Q M/ \M/ \M Qx-B K \ x=5 (Me4N)[Ag(Se5)] 26 /Q\ /Q_Q\ M M M 6 2. High Temperature Metal Chaloogenide Chemistry Metal chalcogenide compounds possess interesting physical properties which can be applied to many uses such as HDS catalysts27, detectors for IR radiation and night vision”, solar cells”, energy storage systems”, and non- linear optical materials31. The great majority of metal chalcogenide compounds known to date have been synthesized at temperatures higher than 500 0C. High temperature conditions are primarily used because the starting materials are often solids and a reasonable diffusion rate between solid reactants is necessary for reaction to take place. During the reaction period, for example, the reactant solids (A and B; the elements of simple binary chalcogenides) partially react to form a layer of product AB at the interface, as shown in the following scheme. _-' - A ‘_ a t— —-l'—'" A AB B l=lt <—— As the layer of product AB becomes thicker, further reaction slows down due to the greater pathlength that A and B must travel (through the AB layer) to react. Thus, in some cases, even though high reaction temperatures are involved, frequent regrindings followed by repeated heating are still necessary to obtain a homogeneous product. Even if a homogeneous product is obtained, growth of single crystals, a necessary prerequisite for proper characterization of new 7 materials, is not readily achieved. To facilitate crystal growth, other techniques such as chemical vapor transport (CVT)32 and molten salts (or flux growth)33 are often employed. CVT works well with binary chalcogenides but often fails with ternary systems and when alkali metals are involved. On the other hand. flux growing is a promising technique and has proven to be successful in a number of binary and ternary chalcogenide systems“. This molten salt technique will be discussed in more detail in the next section. Metal chalcogenide compounds exhibit enormous structural diversity associated with the muIti-bonding characteristics of the chalcogens. For simple binary systems, compounds with MO stoichiometry often adopt either the zinc- blende (e.g. ZnS) or wurtzite-type structure (eg. ZnS) as shown in Figure 1. In both of these structures the metal and chalcogen atoms are tetrahedrally coordinated. Another major structure type found amongst monochalcogenide compounds is the NiAs-type structure as shown in Figure 2 (A). Each O atom is surrounded by a trigonal prism of 6 metal atoms while each metal atom has eight-fold coordination, being surrounded octahedrally by six chalcogen atoms and by two additional metal atoms which are coplanar with four of chalcogen atoms. Furthermore, there are many known defect structures of these major structure types. For example, the two dimensional Ti82 or Cdlz-type structure is a defect NiAs structure. Figure 2 shows the close structural relationships between TiS (NiAs-type) and TiSz (or Cdlz) by removing alternate M atom layers in the TiS structure, which leads to A82 stoichiometry. Another related layer structure is that of M082 where the metal atom has trigonal prismatic geometry and the chalcogens have trigonal pyramidal geometry. (A) Figure1. ORTEP representation of the unit cell of (A) Zinc-blends (ZnS) and (B) Wurtzite (ZnS). Open circles are S atoms and octant-shaded ellipsoids are Zn atoms. (A) Figure 2. ORTEP representation of the unit cell of (A) TiS (NlAs-type) and (B) TI82 (Cdlz-type). Open circles are S atoms and octant-shaded ellipsoids are Ti atoms. 10 Many dichalcogenides have a quite different structural motif, being composed of infinite three-dimensional networks of M and discrete 022' units. The predominate structural types are pyrites (e.g. FeSz) and marcarsite (orthorhombic modifications of FeSz) and are shown in Figure 3. Pyrite can be described as a distorted NaCl-type structure in which 82 units are centered on the Cl positions. Contrary to the well-developed monochalcogenide chemistry, the corresponding chemistry of 03' (x>2) ligands in the solid state is rare and extends only to 022° compounds. Polychalcogenide compounds are thermally unstable at high temperature even in an excess of chalcogen. This instability increases with increasing chain length of polychalcogenide ligands, with the result of smaller chain fragments as shown in eq 2. 0,2- -.--.> QHZ' + nO eq 2. A classical example of such instability is the transformation of TiS(Sz) to Ti32 and S above 500 00.35 Polychalcogenide ligands are more likely to be incorporated into the solid state lattice at lower reaction temperatures. At room temperature few metal polychalcogenide polymers have been crystallized directly from the reaction mixture. These are one-dimensional NH4Cu(S4)35, (Ph4P)2ngTe537, (Me4N)Ag(Qs) (0:838, Seze), (PhaP)Ag(Se4)26(b).39, and two-dimensional (NH4)2Pd(85)(Ss)40. However, room temperature conditions are not particularly conducive to single-crystal growth of compounds with extended structures despite the exceptions mentioned above. It becomes clear that low to intermediate temperatures (150 °C 700 00). However, they have not been used exclusively as reaction media for the synthesis of new materials. Three types of molten salts have been used for recrystallization of known chalcogenide materials. These are (a) metal halides, (b) metal chalcogenides, and (c) alkali metal polysulfides.33(alv43 Among them, alkali metal polysulfide fluxes, or more generally alkali metal polychalcogenide fluxes, are quite attractive for the synthesis of new ternary chalcogenide compounds because they can be used not only as solvents but also as sources of alkali metal and chalcogen. The potential of the polychalcogenide fluxes as reaction media was noted by Scheel about 15 years ago. In his only published paper44 he wrote "The polysulfides of the larger alkali ions are not investigated in detail; they are also potential solvents although the tendency of compound formation increases with ionic radius." However, these fluxes were used only as recrystallization media of known materials at high temperatures. Materials such as ZnS, CdS, MnS, PbS, NaCrSz, KCrS2, NalnS2, KFeS2, F882, NiSz, C082, M032, Nsz, LaS2.x, CU3VS4, and H98 were grown successfully from sodium polysulfide flux at high temperatures (>700 0C) by Scheel44 and others“. An alkali metal polychalcogenide flux can be prepared easily by fusing an alkali metal chalcogenide with stoichiometric amounts of chalcogens above 450 0C as shown in eq 3. A20 + (x-1)Q ------ > AzQx eq 3. One of the most studied systems is the Na28/S flux due to its potential use in sodium/B-aluminalsulfur rechargeable batteries.46 Typical phase diagrams of the polysulfide fluxes (e.g. Na2S/S47 and K2S/S48) are shown in Figure 4. The 14 a. s. 292: m 8 2 8 on 8 8 imuimemui “ _ _ _ c8F _ . .. . _ . n were”; “ coca aew. recon 1 8a a 3. 8m o8 .. 8e. 0 logo max o2. omen on _ 2. assess 89wa as use Smawwnz 3: so assume 82a .4 9:9“. s. 2925 m Tun--- an... on... e do 2: b8 can . 06¢ o 8m 0 com can ease 82 15 local minima in the melting point curves correspond to the eutectic compositions. It is clear that for stoichiometric compositions of Na28x and K28x (x 2 3) the melting points are well below 400 °C and reach 145 0C for K284. Thus, the lowest possible temperature for synthetic chemistry in these system is ca. 145 0C. This temperature lies at the upper limit at which most organic solvents boil off or decompose, yet is low enough to form kinetically stable or metastable products. Similar considerations apply for the other A2Qx systems49 A desirable feature of polychalcogenide fluxes is that upon cooling they solidify to a glassy solid. This glassy matrix can be removed by dissolution in water or nonaqueous solvent and does not interfere with the isolation of the crystalline products. Of course, the products must not dissolve in these solvents. In the corresponding alkali metal polyselenides, the minimum accessible temperature is somewhat higher at 250 oC. Since no other phase diagram information on the heavier chalcogenide (Se, Te) fluxes is available, the reaction temperatures for the A2Q/Q systems are found out in an empirical way. Table 3 contains melting point information on most known alkali metal polychalcogenides. One important property of these alkali metal DOlychalcogenide fluxes is the wide temperature range of their melting points (1 45~1000 0C). This allows great flexibility in the choice of reaction temperature. In addition, the polychalcogenide flux is a highly oxidizing (corrosive) medium as well as a basic (nucleophllic) one. The oxidizing ability allows one to use the elemental form of the metals in a redox reaction. The basiclty of the medium, which can be controlled by varying the A2OIQ ratio, Qovems the reactivity of the polychalcogenide flux and provides the appropriate Iigatnd to the oxidized, positively charged metal. 16 Table 3. Melting Points (00) of Some Known Alkali Metal Polychalcogenides‘w‘49 L123 L123 2 900-975 369.5 N823 N8232 N3233 Na284 N3235 1 180 490 228.8 275 251 .8 Na2Se Na28e2 Na2863 Na28e4 Na28e6 >875 495 313 290 258 Na2Te Na2Te2 (NaaTe2) Na2Te6 953 348 436 K23 K232 K233 K234 K235 K235 840 470 252 1 45 206 1 89 K236 K2362 K2383 K2364 K2365 460 380 205 1 90 K2Te K2T62 K2Te3 R D23 Rb232 FID233 Rb234 R D235 HD233 530 420 213 160 225 201 R bzse R b2Te Rb2Te2 Rb2Te3 C823 C8232 C8233 C8234 C8235 C8236 460 21 7 1 60 21 O 1 86 C8236 Cs2Te 17 Recrystallization of a solid metal sulfide from such fluxes suggests that at some point partial or complete dissolution of the solid metal sulfide occurs through attack by the 83' ion present in solution to form soluble species as shown in eq 4.50 MSn + msxz' _ Nelson?" «14 As with hydrothermal synthesis, a dissolution-reprecipitation mechanism ('mineralizer effect”) is very important and promotes crystal growth. Given the appropriate conditions of temperature and concentration, certain solids, known or new, could crystallize out of solution. 4. New Materials from Polychalcogenide Flux at Intermediate Temperatures Since polychalcogenide fluxes allow easy access to the intermediate temperature region, there are a tremendous number of possibilities for the SYnthesis of new chalcogenide materials using them. As I already mentioned above, this intermediate temperature range can afford new chalcogenide cOrnpounds with the following characteristics: (i) the incorporation of DOchhalcogenide ligands into extended solid state lattices, (ii) metastable or kil’Ietically stable compounds, (iii) low-dimensional structures, and (iv) unique DrOpertieS related to novel structures. Furthermore, this new class of Compounds can bridge the gap between two extremes of chemistry (i.e., solution and solid state). 18 In 1987 lbers and co-workers published the first examples of the use of alkali metal polychalcogenides to synthesize new metal chalcogenide compounds. The compounds, K4Ti381451 and Na2Ti28e352, were obtained from K28x and Na28ex fluxes, respectively, at 375-470 00. These new compounds contain infinite one-dimensional chains of [T i3(82)5(8)2]n4n' and [T i2(8e2)3(8e)2]n2n'. respectively, and are composed of Ti4+ centers bonded to 022' and 02' ligands. Even though they pointed out the potential of this molten salt method for the synthesis of new materials, they did not take advantage of the full potential of these polychalcogenide fluxes and confined their efforts to the relatively higher temperature region (>375 00). Our initial approach for the systematic investigation of new ternary chalcogenide compounds concentrated primarily on lower reaction temperatures (T < 375 00 except for Texz'); we attempted to obtain materials containing even higher polychalcogenide ligands. Indeed, the lower reaction temperature region (215 00 < T < 310 OC) afforded new metal chalcogenide compounds containing 0% (0:8, Se; x=2, 3, 4, and 5) ligands (see Chapter II and Ill and VIII)5°-53. Furthermore, many new structural features with properties ranging from semiconducting to metallic are obtained (see Chapters lV-Vlll).54 In this dissertation, I will describe the synthesis, characterization, and structural Chemistry of new metal chalcogenide compounds prepared by the use of alkali metal polychalcogenide fluxes at intermediate temperature. CHAPTER 2 Synthesis and Characterization of Low-Dimensional (Poly)chalcogenide Compounds In the AlCu/Q System (A=Na, K, Cs; 0:8, Se) 1. Introduction Thus far, the great majority of synthetic metal/chalcogenide chemistry has been carried out either in solution at (or near) room temperature or in the solid state at high temperatures (T > 500 OC). The compounds obtained at low temperatures are soluble, discrete molecular species containing a diverse repertoire of 0x2' (0:8, 8e, Te) ligands with x ranging from 1 (e.g., [Na2F61383013')10 to 9 (e.g., [Au89)]l')ll. By far, among the discrete polychalcogenides, the Q42- ligands occur most frequently. The high temperature compounds tend to be extended, three dimensional, solid state structures containing either 02' or 022' ligands. This is because the DOchhalcogenides are thermally unstable at high temperatures and dissociate into lower polychalcogenides and chalcogen. In general, low-temperature Compounds can be viewed as metastable and thus capable of transforming to their high temperature solid state counterparts via interesting and perhaps isolable low-dimensional intermediates in the intermediate temperature range (1 50-500 oC). Increasing interest in metastable low-dimensional polychalcogenides derives not only from the catalytic27 and electronic55 19 20 properties of these materials but also because they offer a bridge between molecular and solid state chemistry. Most known binary and ternary Cu/S compounds exhibit interesting electronic properties and even mixed-valency associated with Cu1+’2+ ions is often achieved. Thus, our initial investigation concentrated on the ternary AlCu/O systems. In this chapter, we shall illustrate that novel one-dimensional compounds of a-ACuQ4 and p-KCu8453ia) containing tetrachalcogenide (042') ligands, one-dimensional mixed-valence compound of Naacu.48e4, and mixed selenide and oxide phase of Na1,9CU2Se2-CU2O are readily formed in the polychalcogenide flux at intermediate temperatures. 2. Experimental Section 2.1 Reagents Chemicals were used as obtained: copper powder, electrolytic dust, purified, Fisher Scientific Co., Fair Lawn, NJ; copper (ll) oxide (CU20), 97 % purity, Aldrich Chemical 00., Milwaukee, WI; selenium powder, 100 mesh, 99- 95% purity, Aldrich Chemical 00., Milwaukee, WI; sulfur powder, sublimed, J. T- Baker Chemical Co., Phillipsburg, NJ; potassium and sodium metal, analytical reagent, Mallinckrodt lnc., Paris. KY; cesium metal, 99.98% purity, AESAR, Johnson Matthey, Seabrook, NH. 21 2.2. Physical Measurements The FT-IR spectra of the compounds were measured as pellets in a CSI matrix. Each sample was ground with dry Csl into a fine powder and a pressure of about 6 tons was applied to the mixture to make a translucent pellet. The spectrum was recorded in the far IR region (600 to 100 cm”) with the use of a Nicolet 740 FT -IR spectrometer. Variable temperature four-probe dc resistivity and thermoelectric power data for Na1,9CU2Se2-CU2O (5~300 K) and four-probe dc conductivity data for a-KCu8e4 (80~300 K) were provided by Prof. Carl R. Kannewurf (Northwestern University). A computer automated measurement system was employed to obtain thermopower and/or resistivity data with both thermal gradient and/or the current applied along the needle axis of a-KCu8e4 and the (001) plane of Na1,gCU28%CU2O. For all measurements electrode connections to the small single crystals were made with the use of 25 and 60 pm gold wires and gold bonding paste. Quantitative microprobe analysis of the compounds was performed on a Jeol 3SCF scanning electron microscope (SEM) equipped with Tracor Northern TN 5500 X-ray microanalysis attachment. Single crystals of each sample were carefully picked and mounted on an aluminum stub using conducting Silver Paint to help dissipate charges that developed on the sample surface during measurements. Energy Dispersive Spectra (EDS) were obtained using the following experimental set-up: X-ray detector position : 55mm Working distance : 39mm Accelerating voltage : 20 KV Take-off angle : 27 deg 22 Beam current : 200 picoamps Accumulation time : 100 seconds Window : Be A standardless quantitative (SQ) analysis program was used to analyze the X- ray spectra obtained. Since the selenium ratio is always underestimated due to an artifact of the program, a correction factor (x186), which was determined with the known KICu/Se ternary compounds, was used to evaluate the selenium ratio. The analyses reported here are the average of four to six individual measurements on different crystals of each compound. 2.3. Synthesis Chemicals were measured and loaded in Pyrex tubes under a dry nitrogen atmosphere in a Vacuum Atmospheres Dri-Lab glovebox. Potassium sulfide, K28 3.2 g (0.1 mol) of sulfur powder was combined with 7.8 g (0.2 mol) of sliced potassium metal in a round-bottom flask equipped with a Teflon valve and a stirbar. A 100-mL volume of liquid ammonia was condensed into a flask at -78 00 (dry ice/acetone bath) and the mixture stirred for a couple of hours or until the potassium metal had dissolved completely. When a dark blue solution formed, the NH3 was removed by evaporation at room temperature (by allowing the cold bath to warm slowly) under a flow of nitrogen. The resulting light yellow solid was dried in vacuo, flame-dried, and ground to a fine powder in the glovebox. It was used without any further characterization. Preparation of Na2Se and K28e was accomplished by following the procedure used for the K28 preparation with sodium or potassium metal and elemental selenium in a 2:1 ratio. 23 Cesium selenide, CS28e 13.3 g (0.10 mol) of melted cesium metal was put in a round-bottom flask fitted with a Teflon valve by using a pipet. A 100-mL volume of liquid ammonia was condensed into the flask at -78 00 (dry ice/acetone bath), and then 3.9 g (0.05 mol) of elemental selenium and a stirbar were added to the solution. The mixture was stirred for a couple of hours or until the cesium metal had dissolved completely. When a dark brown solution formed, the NH3 was removed by evaporation at room temperature (by allowing the cold bath to warm slowly) under a flow of nitrogen. The resulting orange red solid was dried in vacuo, flame-dried, and ground to a fine powder in the glovebox. It was used without any further characterization. Caution: Direct addition of elemental selenium to cesium metal in a round- bottom flask is dangerous, particularly, if the cesium is not solidified completely. To avoid a spontaneous and violent reaction between the two reactants, we decided to dissolve cesium metal in liquid ammonia first and then to add elemental selenium. tit-Potassium (1,2-p2-tetrasulfldo)cuprate(l), a-KCu84 (I) 0.221 g (2.0 mmol) of K28, 0.064 g (1.0 mmol) of Cu powder, and 0.256 g (8.0 mmol) of 8 powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer- controlled furnace and heated at 215 0C for 72 hrs and cooled slowly to 50 0C at a rate of 2 OCIhr. Orange red needle-like crystals were obtained by removing excess molten potassium polysulfides with water under an N2 atmosphere (yield; 72 % based on the Cu used). A quantitative microprobe analysis performed on a number of crystals with the EDS/SEM system gave an average composition of K1_oCu1,oS4,1. 24 B-Potassium (1,4-p2-tetrasulfldo)cuprate(l), p-KCuSlt (II) 0.221 g (2.0 mmol) of K28, 0.064 g (1.0 mmol) of Cu powder, and 0.256 g (8.0 mmol) of 8 powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer- controlled furnace and heated at 250 0C for 72 hrs and cooled Slowly to 50 °C at a rate of 2 OC/hr. Orange red needle-like crystals were obtained by removing excess potassium polysulfides with water under an N2 atmosphere (yield; 53 % based on the Cu used). A quantitative microprobe analysis performed on a number of crystals with the EDS/SEM system gave an average composition of K1 .OCU1 .0341. tit-Potassium (1,2-p2-tetraselenIdo)cuprate(l), a-KCuSe4 (Ill) 0.157 g (1.0 mmol) of K2Se, 0.032 g (0.5 mmol) of Cu powder, and 0.316 g (4.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum («10'3 torr). The tube was placed in a computer-controlled furnace and heated 250 °C for 96 hrs and cooled slowly to 50 0C at a rate of 2 OC/hr. Dark red needle-shaped crystals were obtained by removing excess potassium polyselenides with water under an N2 atmosphere (63 % based on the Cu used). A quantitative microprobe analysis performed on a number of crystals with the EDS/SEM system gave an average composition of K1.oCut.oSe4.2. a-Ceslum (1,2-p2-tetraselenido)cuprate(l), a-CsCu8e4 (IV) 0.345 g (1.0 mmol) of CS2Se, 0.032 g (0.5 mmol) of Cu powder, and 0.316 g (4.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame sealed under vacuum (~10‘3 torr). The tube was placed in a computer-controlled furnace and heated at 250 °C for 72 hrs and cooled slowly to 50 °C at a rate of 2 OClhr. Black needle-like crystals were obtained in 25 quantitative yield by removing excess molten cesium polyselenides with water under N2 atmosphere (57% based on the Cu used). A quantitative microprobe analysis performed on a number of crystals with the EDS/SEM system gave an average composition of 051 ,oCu1,oSe4,2. Trisodium tetra(p3-selenldo)tetracuprate(I,II), Na3Cu48e4 (V) 0.250 g (2.0 mmol) of Na28e, 0.048 g (0.75 mmol) of Cu powder, and 0.316 g (4.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 350 °C for 96 hrs and cooled slowly to 50 °C at a rate of 2 OCIhr. Black needle-shaped crystals were obtained by removing excess sodium polyselenides with dimethylformamide (DMF) under an N2 atmosphere (yield; 58 % based on the Cu used). However, repeated reactions did not always yield a homogeneous product. The product was often contaminated with a competing phase, Na1,9CU2Se2-CU2O (see below). A quantitative microprobe analysis performed on a number of crystals with the EDS/SEM system gave an average composition of Na3,oCua,98e3_9. Sodium bls(p4-selenldo)dlcuprate(l)-copper oxide, Na1,gCu2 Se2-Cu20 (VI) 0.187 g (1.5 mmol) of Na28e, 0.034 g (0.50 mmol) of Cu powder, and 0.316 g (4.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 340 °C for 96 hrs and slowly cooled to 50 °C at a rate of 2 OCIhr. The black, thin plate crystals were obtained by removing excess sodium polyselenides with DMF under an N2 atmosphere (yield; 76 % based on the Cu used). The product was often contaminated with a competing phase, NaacU4Se4 (see above). 26 Since the product contains oxygen, we have used CU2O instead of Cu metal. 0.187 g (1.5 mmol) of Na2Se, 0.143 g (1.0 mmol) of Cu powder, and 0.316 g (4.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10“3 torr). The tube was placed in a computer-controlled furnace and heated at 330 0C for 7 days and cooled slowly to 50 0C at a rate of 2 OCIhr. Black, thin plate crystals were obtained by removing excess sodium polyselenides with DMF under an N2 atmosphere (yield; 84 % based on the Cu used). The product was also contaminated with Na30u.18e4. 2.4. x-ray Crystallographic Studies All compounds were examined by X-ray powder diffraction for the purpose of phase characterization and identification. The d-spacings for each compound were obtained from the powder patterns recorded on a Phillips XRG- 3000 computer-controlled powder diffractometer, operating at 40KV, 35 mA. Cu-Ka radiation was obtained by use of a graphite monochromator. To verify product homogeneity, d—spacings observed for the bulk materials were compared, and found to be in accord, with the d-spacings calculated from the single crystal X-ray structure analysis data. The d—spacings observed for (V) and (VI) showed extra reflections due to coexisting phase, (VI) and (V), respectively. Calculation of d-spacings was performed using the POWD10 program56. The results are summarized in Tables 4-9. Table 4. Calculated and Observed X-ray Powder Diffraction Pattern of 27 a-KCU34 H K L 6mm) do“ (A) lllmax (obs) 0 1 1 6.94 7.05 100 0 0 2 6.26 6.35 23.9 0 1 2 5.01 5.06 68.1 1 1 0 4.43 4.48 7.0 0 2 1 3.95 3.98 31.9 1 2 0 3.26 3.29 33.9 0 0 4 3.13 3.15 87.0 1 1 3 3.04 3.06 20.1 0 1 4 2.93 2.94 41.1 1 2 2 2.89 2.91 48.9 o 3 1 2.71 2.72 46.4 1 0 4 2.69 2.70 19.3 1 2 3(1 1 4) 2.57 2.58 52.7 o 3 2 2.540 2.55 41.7 2 1 1 2.453 2.466 35.6 1 3 1 2.410 2.420 42.2 2 1 2 2.323 2.323 28.0 2 1 3 2.146 2.155 14.5 0 4 0 2.084 2.093 57.4 1 3 4 1.9332 1.9402 28.5 1 1 6(2 3 1) 1.8908 1.8975 9.3 0 3 5 1.8619 1.8694 16.5 1 2 6 1.7599 1.7623 42.8 2 3 3 1.7352 1.7412 15.5 0 5 1(3 1 2) 1.6530 1.6551 42.8 3 2 0(0 5 2) 1.6123 1.6158 17.1 (2 1 6) 2 4 2 1.5792 1.5823 21.3 28 Table 5. Calculated and Observed X-ray Powder Diffraction Pattern of B-KCuS4 H K L dm(A) dobs (A) lllmax (obs) 0 2 0 8.38 8.52 100 0 1 1 5.94 6.02 67.3 -1 1 2 3.26 3.28 11.9 1 2 1 3.04 3.06 13.4 0 5 1 2.96 2.98 78.4 o 6 1 2.55 2.56 19.9 -2 3 1 2.379 2.369 11.9 -2 3 3 1.9220 1.9174 9.6 -2 8 2 1.5922 1.5918 10.5 Table 6. Calculated and Observed X-ray Powder Diffraction Pattern of a- 29 KCU364 H K L dmb(A) dobs (A) l/lmax (obs) 0 1 1 7.17 7.25 22.8 0 0 2 6.41 6.49 23.1 0 1 2 5.15 5.20 100 0 1 3 3.83 3.87 5.7 1 1 2 3.76 3.79 8.2 1 2 0 3.40 3.42 12.5 1 2 1 3.29 3.31 30.6 0 0 4 3.20 3.23 72.1 1 1 3 3.14 3.17 12.1 o 2 3 3.04 3.06 48.4 o 1 4(1 2 2) 3.00 3.02 61.7 0 3 1 2.81 2.83 19.9 1 0 4 2.77 2.79 19.4 1 2 3 2.66 2.68 42.5 1 1 4(0 3 2) 2.63 2.65 22.6 2 1 1 2.57 2.58 37.7 1 3 1 2.50 2.52 13.1 2 1 2 2.429 2.444 12.9 2 2 1 2.286 2.299 4.0 2 1 3 2.237 2.221 18.9 0 4 0 2.165 2.174 17.4 2 1 4 2.031 2.043 9.4 1 3 4 1.999 2.011 17.9 2 3 1 1.969 1.979 9.9 0 3 5 1.9175 1.931 10.5 2 2 4 1.8818 1.892 3.5 1 3 5 1.8109 1.822 26.9 3 1 2 1.7299 1.738 17.6 2 3 4 1.6926 1.699 30.1 0 5 2 1.6721 1.680 15.1 2 1 6 1.6576 1.667 13.1 2 4 2 1.6452 1.651 12.7 2 4 3 1.5814 1.589 5.3 30 Table 7. Calculated and Observed X-ray Powder Diffraction Pattern of a.- CsCuSea H K L deA) dob, (A) l/lmax (obs) 0 1 1 7.56 7.67 12.6 002 6% 7m $7 0 1 2 5.51 5.56 19.2 1 1 0 4.73 4.77 18.2 1 1 1 4.48 4.54 14.5 0 2 1 4.28 4.32 17.6 1 1 2 3.91 3.94 22.7 022 3m 3m 48 0 0 4 3.48 3.50 75.7 1 2 1 3.39 3.41 50.9 1 1 3 3.31 3.33 31.0 0 1 4 3.24 3.26 100 1 2 2 3.13 3.14 49.3 1 0 4 2.95 2.96 67.3 1 2 3 2.79 2.81 66.7 0 3 2 2.75 2.77 19.6 2 1 1 2.61 2.62 63.0 033 2a 2a 91 2 1 2 2.486 2.493 14.8 1 1 5 2.401 2.409 8.2 0 2 5 2.369 2.376 24.6 2 1 3 2.309 2.315 19.6 0 4 0 2.253 2.262 16.4 1 2 5 2.180 2.135 7.9 1 3 4 2.106 2.115 25.8 1 1 6 2.085 2.091 110 2 3 2 1.960 1.964 81 1 4 3 1.9051 1.9082 76 2 3 3 1.8698 1.8753 10.0 0 2 7 1.8205 1.8241 22.7 225 1mm tmw M5 0 5 1 1.7878 1.7915 131 3 1 2 1.7598 1.7648 261 2 4 1 1.7382 1.7447 29 8 2 4 2 1.6990 1.7039 35 0 1 0 8 1.6620 1.6639 167 2 1 7 1.5938 1.5919 82 31 Table 8. Calculated and Observed X-ray Powder Diffraction Pattern of Na30u118e4 H K L dc,.c(A) dobs (A) l/lmax (obs) 0 2 o 7.60 7.57 60.3 1 1 0 6.65 6.63 100 1 2 0 5.30 5.28 12.9 0 4 0 3.80 3.80 62.2 2 o 0 3.69 3.69 29.7 1 4 o 3.38 3.38 41.0 1 2 1 3.15 3.15 18.1 1 5 0 2.81 2.81 26.6 24oc1n 2% 2% an 0 6 o 2.53 2.53 35.4 3 1 0 2.433 2.432 9.6 2 3 1 2.377 2.373 11.2 0 6 1 2.129 2.125 15.4 3 2 1 2.013 2.013 29.3 0 0 2 1.962 1.958 14.8 1 1 2 1.8820 1.8737 5.9 261 1mm 1mm %4 --r ll. I: Table 9. Calculated and Observed X-ray Powder Diffraction Pattern of Na1,gCU2Se2-CU2O H K L dm.c(A) dobs (A) lllmax (obs) 0 0 2 10.81 10.79 12.2 0 0 4 5.40 5.41 100 0 0 6 3.60 3.61 32.0 1 0 3 3.43 3.44 2.7 1 0 5 2.90 2.91 5.9 0 0 8 2.70 2.71 24.7 1 1 4 2.463 2.463 7.0 1 0 9 2.048 2.047 2.6 1 1 8 1.9340 1.9330 5.1 2 0 2 1.9250 0 012 1.8032 1.8041 32.5 1 011 1.7575 1.7573 3.8 33 The X-ray single crystal data of a-KCuS4. B-KCuS4, a-KCuSe4, and a-CsCuSe4 were collected on a Nicolet P3 four circle diffractometer with graphite monochromated Mo-Ka radiation using the 6-26 scan mode. The data for N83CU4Se4 and N81,gCU2Se2-CU2O were collected on a Rigaku AFC68 diffractometer with graphite monochromated Mo-Ka radiation using the (0'29 scan mode. Accurate unit cell parameters for all compounds were obtained by the least-squares refinement of the 26, (t), x, and 11) values of 20-25 machine- centered reflections. The stability of the experimental setup and crystal integrity were monitored by measuring three standard reflections periodically (every 100~150 reflections) during the data collection period. The intensities did not Show any appreciable decay. Two absorption corrections were applied to the data of all compounds: an empirical absorption correction based on 11: scans for 3 reflections followed by a DIFABS57 correction. The structures of a-KCuS4, p-KCuSa, a-KCuSe4, a-CsCuSe4 were solved with direct methods using SHELXS-8653 and were refined with the SDP59 package of crystallographic programs performed on a VAXstation 2000 computer. Since (I), (III) and (IV) have a non-centrosymmetric Space group (P212121), their enantiomorphs were checked at the final least square cycle. Among them, only a-KCuSe4 gave significant change in the RlRw value from 4.1/5.4 to 3.5/5.0. Based on this, its absolute configuration was determined. The structures of N83CU4364 and Na2CU28e2-CU2O were solved with direct methods using the SHELXS-86 program and were refined with the TEXSAN6o package of crystallographic programs on a VAXstation 3100 computer. The systematic extinction conditions of Na2CU2Se2-CU2O (h+k+l=2n, hko, Okl, hhl) correspond to 8 possible space groups (l4, l-4, l4/m, I422, l4mm, l-42m, l-4m2, and l4/mmm). Among them, the structure was successfully refined in space groups l4/m, l4mm, l-42m, l-42m, and I4/mmm. All refined atomic positions are related to those in space group 34 I4Immm and the relatively lower R/Rw value of 5.3/7.7 was obtained in space group of I4Immm. Therefore, the space group of I4Immm was chosen . The temperature factors of Cu(2) atoms situated on the (0, 1/2, 0) position were large in all refined space groups, implying partial site occupancy of Cu(2) atom. We tried to refine its site occupancy. However, creation of a vacancy on this site caused relatively higher R/Rw value and the automatic refinements gave more than 100 % occupancy (104 %) and even larger temperature factors (7.8 A2). Since 104 % occupancy is physically unrealistic, this site occupancy was fixed at 100 %. Instead, the refinements of Na site occupancy yielded slightly the lower R/Rw value of 5.3/7.7 versus 5.7/7.9 at the final least square cycle. Based on this refinement, the nonstoichiometric formula of Na1,gCU28e2-CU2O was established. All atoms were refined anisotropically. The complete data collection parameters and details of the structure solution and refinement for every compound are given in Table 10. The final coordinates, temperature factors and their estimated standard deviations of all atoms are shown in Tables 11-16. 35 Table 10. Summary of Crystallographic Data for ct-KCuS4, p-KCuS4, ct- KCuSe4, a-CSCuSe4, Na30u.48e4, and Na1,gCu2Se2-CU2O compound I II III Formula a-KCu84 B-KCuS4 a-KCuSe4 Formula weight 230.54 230.54 418.48 space group P212121 P21/C P212121 a (A) 5.245(1) 5.260(2) 5.509(2) b (A) 8.338(3) 16.771 (6) 8.660(3) c(A) 12.539(3) 6.928(5) 12.826(4) 61 (deg) 90.0 90.0 90.0 6 (deg) 90.0 113.52(1) 90.0 y (deg) 90.0 90.0 90.0 Vol (A3), 2 548.4(4), 4 560(1), 4 612.0(3), 4 Temperature (0C) 23 23 -98 Crystal size (mm) 0.4x0.08x0.05 0.4x0.06x0.06 0.4x0.05x0.04 Radiation Mo-Ka Mo-Kct Mo-Ka p. (Mo-Ka, cm“) 60.6 59.3 276.8 Dcalc (gch) 2.80 2.74 4.54 26",” (deg) 49 46 50 Scan method 6/20 6/26 6/20 No. of data collected 1158 917 1322 N0. of unique data 921 823 1064 N0. of data used 877 711 994 (F02>30(F02)) No. of atoms 6 6 6 N0. of variables 55 55 56 Phasing technique Direct methods Direct methods Direct methods Final RlRw 4.9/5.5 3.4/4.2 3.5/5.0 Max. shift/esd 0.0 0.0 0.0 (final cycle) Extinction coefficient NIA NIA 2.96x10'7 Table 10. (cont'd) 36 Formula Formula weight space group a (A) b (A) c (A) 0 (deg) B (069) 11 (deg) Vol (A3), 2 Temperature (0C) Crystal size (mm) Radiation p (MO-K01, cm'1) Dcalc (glcm3) 29rnax (deg) Scan method N0. of data collected N0. of unique data No. of data used (F02>30(F02)) N0. of atoms No. of variables Phasing technique Final RIRw Max. shift/85d (final cycle) Extinction coefficient commnd IV V VI ct-CsCu884 Na30u.4884 Na1_gCu2Se2-CU2O 512.29 638.99 474.08 P212121 Pbam I4Immm 5.571 (2) 7.396(4) 3.914(2) 9.014(4) 15.207(5) 3.914(2) 13.931 (6) 3.924(4) 21.623(4) 90.0 90.0 90.0 90.0 90.0 90.0 90.0 90.0 90.0 699.6(5), 4 441.4(5), 2 331.3(4), 2 23 23 23 0.38X0.05X0.03 0.39x0.03x0.03 0.03x0.16x0.18 MO-Ka MO-Ka MO-Ka 287.1 259.4 236.2 4.86 4.81 4.75 48 60 55 6/20 (”/20 (ll/29 1380 700 254 1 105 700 1 51 965 438 107 6 6 5 56 35 13 Direct methods Direct methods Direct methods 2.7/3.0 3.0/3.6 5.3/7.7 0.0 0.0 0.0 4.60x10'7 NIA NIA 37 Table 11. Fractional Atomic Coordinates and BBQ Values for a-KCu84 with Their Estimated Standard Deviations in Parentheses Atom x y z nga. A2 CU 0.0214(3) 0.3558(2) 0.5680(1 ) 1 .97(2) K 0.3279(6) 0.5696(3) 0.3455(2) 264(5) 3(1) -0.1317(5) 0.3042(3) 0.3978(2) 158(5) 3(2) 0.1243(5) 0.1389(4) 0.3326(2) 164(4) 3(3) 0.0478(5) -0.0745(3) 0.4149(2) 177(5) 3(4) -0.3361(6) ~0. 1092(3) 0.3887(2) 190(5) 5' 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as ng = (413)[82811 + b2822 + 02533 + ablcos 10812 + aclcos 0813 + bclcos (1)523]- Table 12. Fractional Atomic Coordinates and Beq Values for 8-KCuS4 with Their Estimated Standard Deviations in Parentheses Atom x y z nga, A2 CU 0.2456(2) 0.49577(6) 0.4965(2) 2. 22(2) K 0.2754(4) 0.6199(1) 0.0476(3) 2. 67(4) 3(1 ) 0.0695(4) 0.3985(1 ) 0.6585(3) 1 .95(4) 3(2) -0.0991 (4) 0.1883(1) -0.0715(3) 215(4) 3(3) 0.5195(4) 0.6803(1) 0.6497(4) 272(5) 8(4) 0.3777(4) 0.4359(1) 0.2423(3) 1.91(4) a B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[82311 + b2822 + 02333 + ablcos 10812 + acloos BlBta + bclcos 00823]. 38 Table 13. Fractional Atomic Coordinates and ng Values for a-KCuSe4 with Their Estimated Standard Deviations in Parentheses Atom x y z nga, A2 Se(1) 1.1417(3) 0.6855(2) 0.6029(1) 042(2) 36(2 ) 0.8567(3) 0.8596(2) 0.6782(1) 0.41 (2) 88(3) 0.9272(3) 1.0848(2) 0.5762(1) 040(3) 36(4) 1 3440(3) 1 .1 1 98(2) 0.6076(1 ) 050(3) Cu 0.9733(4) 0.6422(2) 0.4301(2) 059(3) K 0.6816(8) 0.4418(4) 0.6579(3) 119(7) a E values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Baq = (4/3)[aZB11 + 02822 + 02833 + ablcos 10812 + £1de 10813 + bclcos (1)823]- Table 14. Fractional Atomic Coordinates and ng Values for a-CsCuSe4 with Their Estimated Standard Deviations in Parentheses Atom x y z Baqa, A2 CS 0.3330(2) 0.56051 (9) 0.33090(6) 236(2) 38( 1) -0. 1351 (2) 0.3034(1) 0.40335(9) 1.71 (2) 38( 2) 0.1543(2) 0.1385(1) 0.33804(9) 1 .78(2) 36(3) 0.0861 (2) -0.0798(2) 0.43006(9) 1 .99(2) 36(4) -0.3281 (2) -0.1 150(1) 0.40809(9) 1 .91 (2) CU 0.0414(3) 0.3634(2) 0.5587(1) 214(3) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as ng = (4/3)[a2811 + b2822 + 02833 + ablcos 10812 + ac(oos 19813 + bclcos a)823l 39 Table 15. Fractional Atomic Coordinates and ng Values for Nanu4Se4 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 36(1) 0.2720(1) 0.153340) 0 1 26(4) Se(2) 0.7472(1) 0.089440) 1/2 1.26(4) CU(1) 0.5694(2) 0.0899(1) 0 1.73(6) CU(2) 0.3461(2) 0.0647(1) 1/2 1.93(6) Na(1) 0 0 1 .0000 1 .8(3) Na(2) 0.5103(6) 0.2629(3) 1 I2 2.0(2) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[aZB11 + b2822 + 02833 + ab(cos 10812 + £1de BlBts + bc(cos (1)323]- Table 16. Fractional Atomic Coordinates and Beq Values for Na1,QCU2Se2CU2O with Their Estimated Standard Deviations in Parentheses Atom x y z nga, A2 36 0 0 0.1793(2) 1 .45(9) CU(1) 1/2 0 1/4 2.0(1) Cu(2) 0 -1/2 0 6.1(2) Na 1/2 -1/2 0.118(1) 2.9(5)b 0 o 0 0 0.7(7) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as ng = (4/3)[a2811 + b2822 + 02833 + ab(cos y)B12 + ac(cos (3)813 + bc(cos 00323]. b The refined occupancy of Na site is 96 (5) %. 4O 3. Results and Discussion 3.1. Synthesis and Spectroscopy The synthesis of all compounds has been readily accomplished at the intermediate temperature range (215 OC - 400 00) using alkali metal polychalcogenide fluxes as solvents and reagents. Each reaction formed a complete melt at the temperature employed. Alkali metal polychalcogenide flux is easily prepared by fusing alkali metal monochalcogenide with stoichiometric amounts of elemental chalcogen as shown in eq 1. A20 + (1-x)Q ------> A2Qx (A=Na, K, Cs; 0:8, Se) eq 1. The highly reactive and corrosive A2Qxfluxes at the employed temperatures oxidize Cu metal and are reduced to shorter polychalcogenides as shown in eq 2. AQQX + Cu ------ > AaCuqu + A2Qy (1sy 500 oC). The reaction of Cu in polysulfide fluxes (e.g. K2Sx, Na28x) at 375 °C or above yielded several known compounds such as KCU48361, Cu862, and Na;)CU4S463 which had been 41 prepared at relatively higher temperatures. This was disappointing, but it suggested that these temperatures were still too high to stabilize any new phases in this system. Therefore, we lowered the reaction temperature to 250 °C and isolated a novel polychalcogenide phase, p-KCu84, using polychalcogenide flux (e.g. K285). When we used a polysulfide flux with shorter chains (e.g. K283) the thermodynamically stable phase, KCU48361, was still isolated at 250 °C . This is probably due to the higher reactivity of this more basic shorter chain flux. When the reaction temperature was lowered to 215 OC, surprisingly, another polychalcogenide phase, a-KCUS4, was obtained using the same reactant ratio K28/Cu/S 012/1I8 (e.g. in K285 flux) as that of p-KCu84. Since the a- and 8-phase are formed from the same reactant ratio but at slightly different temperatures, they are expected to originate from the same solution precursor. The B-phase, which requires a slightly higher temperature to form, is probably more thermodynamically stable than the cit-phase. The a- to B- transformation is not reversible and was only observed in the polysulfide flux (K285). Attempts to effect an a- to B-conversion by heating crystals of a-KCu84 at 250 0C in the absence of K285 flux resulted in decomposition to CuS and K2Sx. These observations underscore the importance of the nature of the polychalcogenide flux and the reaction temperature in stabilizing new metastable phases. They emphasize the validity of the intermediate temperature approach that we have undertaken. The extension of the above reaction into polyselenide systems showed chemistry similar to that seen in polysulfide system. a-KCuSe4 was obtained in the polyselenide fluxes (K2Sea-vK2Se9) at 250 oC. Since p-KCuS4 was formed at slightly higher temperature compared to the low-temperature a-polymorph, we increased the reaction temperature to 390 °C to obtain the Se-analogue of p-KCu84. However, a-KCu8e4 is the only phase we have seen in the entire 42 temperature region (250-390 0C). The expected p—KCu8e4 perhaps requires a higher temperature to overcome the activation barrier of the a- to p-conversion or it can not exist as a stable lattice. During this study we saw another promising phase in the reactant ratio K2SeICuISe of 1I1/8 at 250 0C, but we could not reproduce it for sometime. It has the approximate composition of KCuSey as determined by the EDS/SEM analysis. Since this phase has a composition more selenium-rich than a- KCuSe4, we increased the selenium ratio from 1/1/8 to 1/1/16. Of these, 1/1I12~13 ratios yielded better crystals at 350 0C. The preliminary X-ray single crystal study was unsuccessful due to poor crystallinity and hair-like morphology. We were unsuccessful in refining the structure, but preliminary refinements showed highly disordered 8862' and 8e42' units. Recollection of the data is necessary to complete the structural refinement. The cell parameters of this new phase are: a=12.560(4) A, b=5.449(6) A, c=22.949(4) A, and 6:92.95(2) deg. Work to obtain better single crystals is in progress. The use of cesium polyselenide fluxes (C82363~C82369) yielded results similar to those seen in potassium polyselenide flux. a-CsCu8e4 was the only product obtained in the CSlCu/Se system in the temperature range of 250- 350 0C. The use of Na28ex flux, which requires a slightly higher reaction temperature to form melt, did not afford any polychalcogenide compounds. Two competing phases, Na30u218e4 and Na1,gCU2Se2-CU2O, were isolated. Even though we suspected that either Na283 or Cu metal had become contaminated with oxygen, isolation of oxygen containing phase was quite surprising to us because of the well known Cu chalcophilicity. The Na2Se is likely the oxygen source in the reaction due to the surface oxidation of sodium metal producing NaOH during the period of its preparation. Many variations in the reaction 43 conditions were made to obtain pure NagCU4Se4 and Na1gCU2Se2-CU2O phases. We used fresh Na2Se every time and varied reactant ratios and reaction temperatures, but Na2CU2Se2-CU2O was always present as a coexisting phase. Among them, only the reactant ratio Na2Se/Cu/8e of 4/1.5I8 at 330 00 gave a relatively homogeneous product of Na3CU4Se4. The rest of the reactions yielded N81,9CU2Se2-CU2O as the major product with a small amount of NaaCU4364 which can be identified by its X-ray powder diffraction (XRD) pattern. Especially the 3/1/8 ratio at 340 00 yielded a relatively pure phase of Na1_gCU2Se2-CU2O with large (1 mm) single crystals, but it was still contaminated with Na:;CU4Se4. When we increased the reaction temperature to 450 00, increased amounts of the N81,9CU28e2-CU2O phase were observed, implying that it is thermodynamically the more stable phase. Preparation of Na1,QCU2Se2CU2O using CU2O instead of Cu metal at 330 0C gave almost the same results we had seen in the Na2Se/Cu/Se system. The reactant ratio Na2Se/CU2O/Se of 3/2/8 yielded fairly pure Na1,9CU2Se2-CU2O with small contamination of Na3CU48e4. We also used a different source of Na28e, purchased from CERAC. However, it yielded discouraging results. The reaction mixtures did not form melts even at 450 °C and selenium simply evaporated into the colder regions of the reaction tubes, probably because the particle size (-100 mesh) of Na2Se purchased was larger than that which we prepared or, more likely, the Na2Se was grossly impure. Further work is in progress to obtain homogeneous phases and large size single crystals suitable for charge transport property measurements. In the far-IR region (I), (II), (III), and (IV) exhibit spectral absorptions due to O-Q and/or Cu-Q stretching vibrations as Shown in Figure 5. Observed absorption frequencies are given in Table 17. 44 Figure 5. Far-IR spectra of (A) a-KCu84, (B) B-KCu84, (C) ot-KCuSe4, and (D) a-CSCUSe4. T (96) T (95) 45 (A) (B) (C) (D) 198 ton") 350 295 I 240 Figure 5. 185 130 1cm") . 46 Table 17. Frequencies (cm'i) of Spectral Absorptions of (I), (II), (III), and (IV) due to Q-Q and M-Q Stretching Vlbrations (I) (II) (III) (IV) a-KCuS4 p-KCuS4 a-KCuSe4 a-CsCuSe4 481 (3)8 481 (m) 249 (s) 252 (s) 435 (m) 456 (s) 241 (w) 246 (s) 412 (m) 436 (m) 180 (m) 236 (w) 286 (m) 412 (m) 158 (w) 179 (m) 259 (m) 269 (w) 145 (m) 175 (w) a (s) strong; (m) medium; (w) weak The polysulfide compounds in this study, a- and p-KCuS4, show a common absorption at 481 cm'1 which is attributed to a 8-8 stretching vibration by comparison with the spectra of other known polysulfide compounds.64 The vs.s vibrational frequencies are observed in the range of 446~500 cm'1. The additional common peak at 435 cm'1 can be attributed to a Cu-S stretching vibration. For Cu/Se42' polyselenide compounds, a spectral absorption is observed in the range of 246~252 cm". This band can be assigned to Se-Se stretching vibrations by comparison with the spectra of other known polyselenide complexes and with that of the unbound ligand (Ph4P)288526(b) «39-39 267 cm'l). In addition, the 1739-33 absorption in this region has been observed previously in various compounds, eg Se(2'65 (x=1-6) at 258 cm'l, 083655 at 253cm'1, [Fe2881212'24 at 258cm'1, and [Pd(Se4)2}2'67 at 274cm'l. 47 In CulSe42' compounds an additional absorption band was found in the vicinity of 240 cm'1. This can be a possible candidate for a Cu-Se stretching vibration. It is usually difficult to interpret the IR spectra of metal polyselenide compounds unambiguously because M-Se and Se-Se stretching frequencies fall in the same low-frequency IR region (200-340 cm'1 for Se) and systematic IR spectroscopic data for the various free ligands (sz'. =2-6) and metal chalcogenide complexes are still lacking. 3.2. Description of Structures 3.2.1. Structure of a-KCuS4(l), a-KCuSe4(lll), and a-CsCuSe4(IV) (I), (II), and (Ill) possess one-dimensional structures as shown in Figure 6. They are isostructural with the known (NH4)Cu8436 and contain tetrachalcogenide ligands (842' or 3642') chelating and also bridging Cu atoms. The infinite noncentrosymmetric [Cu04lnn' (hereafter 0:8, 8e) chains are running parallel to the crystallographic a-axis. The chains are composed of condensed Cu04 five-membered rings related to each other by a crystallographic 2-fold screw axis, parallel to the chain direction as shown in Figure 7. The whole ot-[CU(Q4)]n”' chain can be viewed as an array of fused Cu04 and CU203 five-membered rings. The 042' ligand features an asymmetric bridging mode bound to three Cu+ centers. It chelates one Cu atom via its terminal atoms (0(1) and (0(4)) and bridges two neighboring copper atoms via a terminal atom (0(1)) and an adjacent internal atom (0(2)). The Cu atoms are all crystallographically equivalent and feature a distorted tetrahedral geometry. The average Cu-0 bond distances are 234(6) A, 245(6) A, and 247(8) A in (I), (II), and (Ill) respectively. However, the Cu-0 bridging bond of ' the internal chalcogenide atom (0(1)) is slightly elongated compared to the 48 I a 4 '3 C‘ rt ‘r‘l‘.““ . ‘ ’ ‘ ,vr ! (3| f it???“ "at"??? . 1419 1.1393- fluifl“ . - -~'.~-:."-.=='-;;-:.--"" ATM ' .4 r I l 0.11:: D 0’ .o . \1 (.1 o ’ t g I r 7 ‘1. '2 ‘7 'o ’o 4 4 Q . 6‘ i. "‘23" d\\v{-‘ ':' 9%.," . v 5' \.‘e I. o. O k ‘ . .. 5“ t . a" O 13’; “3’“: 1".“ ’1 C" T k... ’1 ”’0‘ a?) ‘5' . 4 s O A) . -.wc mEQm __mv=w och .6393 6 =8 E: 65 6 26.69% .m 059“. 56 .oEocom 9.538. 5;, Ease .=:_vm:o_-n 2:55 .mco_mcoE_u-oco 05 .o coszomoEo. awhmo .m 659“. o o o o a 4 . a . . ., O .l .0 I\ 0 .. o 3.5 \. 2va o e an; :5 57 Table 21. Selected Bond Distances (A) and Angles (deg) in B-KCuS4 with Standard Deviations in Parentheses Cu-S(1) 2.368(2) S(1)-Cu-S(1) 1 1328(5) Cu-S(1) 2.372(2) S(1)-Cu-S(4) 1 1033(6) Cu-S(4) 2.360(2) S(1)-Cu-S(4) 109.65(6) (Du-S(4) 2.378(2) S(1)-Cu-S(4) 109.95(6) Cu-S (mean) 2.370(7) S(1)-Cu-S(4) 101.75(6) S(4)-Cu-S(4) 111.65(5) S(1)-S(2) 2.078(2) Cu--S(1)-Cu 6673(5) S(2)-S(3) 2.060(2) Cu--S(1)-S(2) 104.32(8) S(3)-S(4) 2.080(2) Cu--S(1)-S(2) 9974(8) S-S (mean) 207(1) S(1)-S(2)-S(3) 103.94(9) S(2)-S(3)-S(4) 104.69(9) K-S(1) 3.237(2) Cu-S(4)-Cu 6835(5) K-S(1) 3.243(2) Cu-S(4)-S(3) 104.02(8) K-S(2) 3.370(2) Cu-S(4)-S(3) 9920(8) K-S(2) 3.259(2) K-S(4) 3326(2) K-S(4) 3.347(2) Cu-Cu 2.607(1) K-S(4) 3.342(2) Cu-Cu 2.661(1) K-S (mean) 330(6) 58 3.2.3. Structure of Naacu4Se4 (V) The compound Na3CU4Se4 possesses a one-dimensional structure as shown in Figure 10. It is isostructural to the known mixed-valence compound Na3CU4S463. The structure of Na3CU4Se4 consists of infinite centrosymmetric [CU4Se41n3n' columns running parallel to the crystallographic o-axis. The center of symmetry is lying at the center of the column. The columns are composed of parallel CU4Se4 eight-membered rings which are connected to each other by the inter-ring Cu-Se bonds as shown in Figure 11. Alternatively, this can be viewed as an array of fused CU3Se3 six-membered rings. Na3CU4Se4 is formally a mixed-valence compound. Based on the formula Na3CU4Se4 and the assumption that the formal oxidation state on each Se atom is -2, the formal oxidation state on Cu atoms is +1 (three atoms) and +2 (one atom). However, the coexistence of Cu2+ and Se2' is thermodynamically unstable with respect to electron transfer from the reducing Se2' to the oxidizing Cu2+. if one then considers all copper atoms in the +1 oxidation state, one electron vacancy exists per four Se atoms. In NQCU4Se4 there are two crystallographically distinct Cu atoms in the asymmetric unit. They feature slightly distorted trigonal planar geometry. The average Cu-Se bond distances, 238(2) A for Cu(1) and 2.443(1) A for Cu(2), are similar to those of three- coordinated Cu“ atoms in TlCuase273 at 2.46(1)~2.47(8) A. There are two crystallographically distinct Se atoms in the asymmetric unit. They are bonded to three Cu atoms with trigonal pyramidal geometry. The average Cu-Se(1) distance of 243(2) A is very similar to that of Cu-Se(2) of 239(5) A. Therefore, these two crystallographically distinct selenium atoms are expected to have indistinguishable oxidation states. Since there is no short Se-Se contacts, we can expect one electron deficiency (hole) to be delocalized on the valence 59 Figure 10. ORTEP representation of the unit cell of Na3CU4Se4. 6O Flgure 11. Two views of the one-dimensional stmcture of [CU4SG4ln3n’ with labeling scheme. 61 band, consisting mainly of p orbitals of four atoms. This gives rise to -1.75 average formal charge on Se. Therefore, Na30u.(Se4 can be reasonably formulated as (Na+)3(Cu+)4(Se1-75')4. These are only formal charges; the actual description is more complex due to the small electronegativity difference between Cu and Se atoms. Based on this partially empty valence band, we expect p—type metallic behavior for this material. There are short Cu-Cu contacts at 2.593(2) A in the column, slightly shorter than those observed in N830U4S4 at 2.619(1) A. These short Cu-Cu contacts can be considered as d10-d1o attractive interactions“. Selected bond distances and angles are given in Table 22. There are two crystallographically distinct Na atoms located between the chains. They participate in ionic interactions with Se atoms and have distorted octahedral geometry with 6 adjacent Se atoms. The shortest Na-Se bond distances are 3.033(2) A for Na(1)-Se(2) and 2.971(5) A for Na(2)-Se(2). 62 Table 22. Selected Bond Distances (A) and Angles (deg) in Na3CU4Se4 with Standard Deviations in Parentheses Cu(1)-Se(1) 2.402(2) Se(1)-Cu(1)-Se(2) 120.73(5) (x2) Cu(1)-Se(2) 2.362(2) (x2) Se(2)-Cu(1)-Se(2) 1 1236(9) Cu-Se (mean) 238(2) Se(1)-Cu(2)-Se(1) 106.89(9) Se(1)-Cu(2)-Se(2) 1 1777(5) (x2) Cu(2)-Se(1 ) 2.443(2) (x2) Cu(1 )-Se(1)-Cu(2) 6473(5) (x2) Cu(2)-Se(2) 2.444(2) Cu(2)-Se(1 )-Cu(2) 1 0689(9) Cu-Se (mean) 2.443(6) Cu(1)-Se(2)-Cu(1) 1 1236(9) Cu(1)-Se(2)-Cu(2) 81 .14(5) (x2) Na(1)—Se(2) 3.033(2) (x4) Na(2)-Se(1 ) 3.120(4) (x2) Na(1)-Se(1 ) 3.080(1) (x2) Na(2)-Se(1) 3.036(4) (x2) Na(1)-Se (mean) 305(2) Na(2)-Se(2) 2.971 (5) Na(2)-Se(2) 3.167(5) Cu(2)-Cu(1) 2.593(2) Na(2)-Se (mean) 808(7) Cu(1)-Cu(1) 2.922(3) Cu(2)-Cu(2) 3.009(3) Cu(1)-Cu(2) 3.126(2) 63 3.2.4. Structure of N31,90u28e2-Cu20 (VI) The structure of Na1,9Cuzse2-CU2O is intriguing and consists of two independent layers of [CuSelnn' and [Cuzo] as shown in Figure 12. It can be considered an intergrowth of NaCuSe74 with a CuzO layer. Figure 13 shows views of the two individual layers. The [CuSelnn' in the NaCuSe fragment is an anti PbO-type layer, where Cu (on -4m2 site) has tetrahedral geometry and Se (on 4mm site) has square pyramidal geometry. The average Cu-Se bond distance of 2.484 (3) is slightly smaller than that found in NaCuSe (2.55 A), but is very similar to those found in a-ACuSe4 (A=K, Cs) (see Tables 19 and 20). The Cuzo layer is a new structural type. In this layer Cu, situated on a crystallographic mmm site, has linear coordination and oxygen on the crystallographic 4/mmm site are bonded to four linear Cu atoms with square planar geometry. Thus, the Cuzo layer can be considered as an anti Cqu- type layer, common in high Tc cuprate superconductors75. Square planar geometry of oxygen is rare in metal oxide chemistry. One known example is Nb076, which contains square planar Nb and O atoms in the framework. The average Cu-O bond distance of linear Cu atoms of 1.957 (1) A is in the normal range found in known Cu/O compounds (1 .85~1.98 A). There are short Cu--Cu contacts in the layers in the range of 2.768(2) A, but no Cu--Cu contacts between the layers. Selected bond distances and angles are given in Table 23. Na1,9Cque20U20 is the first example of coorystallization of a Cu/O layer with 00K) layer. It is interesting to see the similarity in lattice size of two independent layers; this could be the driving force to stabilize the whole lattice. 64 Figure 12. ORTEP representation of the unit cell of Na1,9002382-CU2O with labeling scheme. 65 .mEofi 0 new ow 2a «.926 :30 new 920% :0 2m mu_oma___m 885-2860 .8596 .62 :38 Bass .28. 090 av 62.... 285. E .6 838852 $55 .9659”. a: 2. “I “I “I e n. “n. n. 4.. A. 66 Table 23. Selected Bond Distances (A) and Angles (deg) in NamCque2Cuzo with Standard Deviations in Parentheses Cu(1)-Se 2.484(3) (x4) Se-Cu(1)-Se 104.0(2) (x3) Cu(1)-Cu(1) 2.768(2) Se—Cu(1)-Se 1 12.29(9)(x3) Cu(1)-Se-Cu(1) 67.71 (9) (x4) Cu(1 )-Se-Cu( 1) 104.0(2) (x2) Cu(2)-O 1 957(1) O-Cu(2)-O 1 80.00 Cu(2)-Cu(2) 2.768(2) Cu(2)-O-Cu(2) 90.00 (x4) Cu(2)-O—Cu(2) 180.00 (x2) Cu-Cu Na-Se 3.070(9) (x4) Cu(1)-Cu(1) 3.914(2) Cu(2)-Cu(2) 3.914(2) It should be noted that the temperature factor of the Cu(2) atom in the Cuzo layer is very large (see Table 16). This may suggest positional disorder of this atom on the (0.1/2,0) position. Cu atoms may move back and forth along a- and b—axis to make Cu-O and Cu-Cu distances short and long as shown in the following Scheme 1. This could occur randomly in the lattice and would not be detectable by an X-ray single crystal study, but could be manifested as higher temperature factors for Cu. 67 (I :C>Cu(2) Scheme 1 Based on the structural formula Na1,90que2Cuzo, one could expect p- type metallic, behavior for this material. Partial removal of electropositive elements from the lattice is accompanied with the creation of electron vacancies in the framework (e.g. oxidation of the framework), resulting in a partially empty valence band. This p-type metallic behavior is consistent with the measured charge transport properties (see below). Alternatively, the removal of Cu atoms from the framework also creates electron holes, resulting p-type metallic behavior. However, based on the refinement result, this does not appear to occur to a significant extent. There is only one crystallographically distinct Na atom situated on a crystallographic 4mm site. Na atoms interact ionically only with Se atoms at 3.070 (9) A. They have square pyramidal geometry. 68 3.3. Close Structural Relationships between a- and B- [Cu(Sdlnfi' Chalns. The a-[Cu(S4)]n"' chain is quite distinct from the 6-[Cu(S4)]n"' chain in binding mode of the tetrasulfide ligand, even though the two have a very close structural relationship. For instance, both chains are composed of tetrasulfide ligands and CuS4 five-membered rings. Yet, the bridging mode of tetrasulfide ligand and the Cu atom arrangement in both chains exhibit a significant difference. In the a-[Cu(S4)]n ' chain, the bridging atoms of tetrasulfide ligand are one a terminal atom and the other an internal atom, which lead to the zigzag array of Cu atoms along the chain direction. The bridging atoms in the 6- [Cu(S4)]n"' chain are terminal sulfur atoms of the tetrasulfide. This leads to formation of Cuzszfour-membered rings with a linear array of Cu atoms along the chain direction. If a- and 6-[Cu(S4)]n"' chains are to be formed by the same solution precursor, one can easily visualize the a- to 6-phase conversion as shown in Figure 14. The long Cu-S(2) bond of bridging internal sulfur atoms in ct- [Cu(S4)]n"‘ chain can be easily broken by thermal activation to form the hypothetical intermediate [Cu(S4)]n"' chain. This hypothetical [Cu(S4)]n"' chain is similar in structure to the known [AgSe4]nn-26lb).39 in which Ag+ is trigonal planar. This [AgSe4lnn' chain is also shown in Figure 14 (C), for the comparison with hypothetical intermediate [Cu(S4)]n"' chain. From this hypothetical [Cu(S4)]n"' chain, if S(4) atoms and three-coordinate Cu atoms are brought close to each other, B-[Cu(S4)]n"' chain can be formed. 69 Figure 14. Schematic representation of structural transformation of (A) the a- [CuS4]n"' chain to (D) the B-[CuS4lnn' chain (bottom) via (B) hypothetical [CuS4lnn' intermediate which is similar in structure to (C) the known [AgSe4]n"' chain. (B) ' the a- . . . etical . . e4]n”' . . Figure 14. 71 3.4. Charge Transport Properties of a-KCuSe4 (Ill) and N81,9002302-CU2O (VI) Based on the completely filled valence band of a-KCuSe4 due to the d10 electronic configuration of Cu atoms, one could expect semiconducting behavior of this material. The conductivity data of its single crystal over the temperature range of 80330 K shows increasing conductivity with increasing temperature, which is typical behavior of semiconductors, as shown in Figure 15. However, its values are relatively small and reach 10'6 Slcm at 330 K. These materials are expected to have narrow band width due to their low dimensionality and poor overlaps between Cu and Se atoms between chains. Based on the partial cation vacancy, Na1,9Cque2-Cuzo is expected to be a p-type metallic conductor. The charge transport measurements on single crystals of Na1,90que2-CU20 along the (001) plane show that the resistivity at first decreases linearly with decreasing temperature and at low temperatures levels off to a constant value (so called residual resistivity) over the temperature range 5300 K as shown in Figure 16. This is typical metallic behavior. The resistivity increases from 5x10“6 Q-cm at 5 K to 1.1x10'4 Q-cm at room temperature. There is a little jump in the resistivity around 270 K. Since the compound contains Na ions which are easily hydrolyzed, this may originate from the melting of the water vapor which must have been condensed during the period of measurements. The temperature dependence of the thermoelectric power (Seebeck coefficient) shows a very small positive value of 0.5~3 pV/K in the temperature range of 100~33O K as shown in Figure 17. The small and linearly increasing Seebeck coefficient with rising temperature indicates that NamCquez-Cuzo is a p-type metal. 72 .vomaoxé .338 065m m .2 929352 .o 5:82 m we. Emu 9-509 53:03:06 :85on Boa Son. .9 059”. O: 222883 omm com omm com of 2: cm . _ _ q — q q _ A — a q q — u _ q _ — d . q _ — q u — u PFI O .. (we/s) o 601 73 .O~:0.~omuaoa..az .o .238 0.9.3 o .o. 6.296an. .o 5:25. a am Sun .5903 3.25.3. one... Son. .3 05?. 0: 928.32.”; com omm com om? cop om o _.+J—1—-——4_qu4qddd—-_q-—-qd o CO CG) P our o: cop (mo-Uni) AllAllSlseH 74 6030030200402 .0 .295 0.050 0 .2 0.20.0050. .0 3.6.2 0 00 0.00 $.35 .038 2.80.0259... .2 0.39“. 0.. 0.20.0058. 000 com 000 com om. 8. on ._-q—fiq_l-——~__—.uq44.--—-_-_1000 68%} mm... 1 H U. - m 4.0.? w - O H w mm. M , 0.. . . .m; -om M H (um (.m.m .0, - . o_ co 75 The mixed-valence compound Na3CU4Se4 is also expected to have interesting electronic properties because one- and two-dimensional metallic conductors are often subject to metal-insulator transitions due to their electronic instability". Furthermore, a recent theoretical study63lc) on the sulfide analogue of NaCuSe.) suggested it might be electronically unstable to show as charge density waves (CDW) or spin density waves (SDW). However, Na30u.tS4 is a normal metallic conductor without any anomalies in resistivity and charge transport propertieswb). Therefore, it would be quite interesting to study electronic properties of Na3CU4Se4 to look for such anomalies. Unfortunately, due to the sample inhomogeneity and small size of single crystals, we have not been able to pursue further investigations on its electronic properties. Work to obtain pure materials and large crystals is still in progress.78 We have demonstrated that synthesis at intermediate temperatures (215 to 350 0C) can provide an avenue to metastable polychalcogenide materials and low-dimensional chalcogenide compounds with interesting electrical properties. The structural diversity of 042' ligands, which is pervasive in discrete molecular polychalcogenide compounds, highlighted the potential use of polychalcogenide fluxes at intermediate temperatures in the study of the polychalcogenide ligation and structural diversity in extended structures. CHAPTER 3 Low-Dlmenslonal Compounds Incorporating (Poly)chalcogenide ngands In the AlAu/O (A=Na, K, Cs; 0:8, Se) Systems 1. Introduction Solid state compounds incorporating the versatile polychalcogenide ligands (0x20 are rare. These ligands are thermally unstable and release chalcogen at the high temperatures commonly used in solid state reactions. We believe that an enormous number of interesting polychalcogenide compounds occur at intermediate (e.g. 150-500 °C) temperatures and could be discovered, provided suitable solvents (e.g. molten salts) were available. These temperatures do not support the conventional (non)aqueous solvents commonly used near room temperature, nor do they favor true solid state reactions due to exceedingly slow diffusion rate. This has been particularly true in chalcogenide chemistry.79 We have been successful in synthesizing new polychalcogenide compounds of a- and B-ACuQ4 using alkali metal polychalcogenide fluxes as solvents and reagents at intermediate temperatures.53(3) (see chapter 2) They contains polychalcogenide ligands (S42' and 6e42') with one-dimensional chain structures. We also have initiated investigations of ternary A/AuIQ systems (A=alkali metal; Q=S, Se) using alkali metal polychalcogenide (AzQx) 76 77 fluxes as solvents and reagents at intermediate temperatures. in the ternary AIAuIQ systems (A: alkali metal) there are only three known compounds of KA u 080 (0:8, Se), N 33A u S 281, and K4A u 68532, containing monochalcogenide ligands. in this chapter, we illustrate the structural diversity associated with various polychalcogenides from 8332' to Se52‘ and the interesting Au‘+’3+ redox chemistry in A2Sex flux. Novel low-dimensional (poly)chalcogenide compounds of KAuQs (Q=S, Se53lcl), CsAuSea, KaAuSe1353lG), Na3AuSe3, and AAqu (A=Na, K) are readily obtained in A20, fluxes at 250-350 °C. 2. Experimental Section 2.1 Reagents Chemicals in this work were used as obtained: gold powder, -325 mesh, 99.95% purity, Cerac, Milwaukee, WI; selenium powder, -100 mesh, 99.95% purity, Aldrich Chemical 00., Milwaukee, WI; sulfur powder, sublimed, J. T. Baker Chemical Co., Phillipsburg, NJ; potassium and sodium metal, analytical reagent, Mallinckrodt lnc., Paris, KY; cesium metal, 99.98% purity, AESAR, Johnson Matthey, Seabrook, NH. 2.2. Physical Measurements FT -IR spectra were determined as a pellet in a Csl matrix. Each sample was ground with dry Csl into a fine powder and a pressure of about 6 tons was applied to the mixture to make a translucent pellet. Spectra were recorded in 78 the far IR region (600 to 100 cm'l) with the use of a Nicolet 740 FT-IR spectrometer. Quantitative microprobe analysis was performed on a Jeoi 35CF scanning electron microscopy equipped with Tracor Nothern TN5500 X-ray microanalysis attachment. Single crystals of each sample were carefully picked and mounted on an aluminum stub using conducting silver paint to help dissipate charges that develop on the sample surface during measurements. Energy Dispersive Spectra (EDS) were obtained using the following experimental set-up: X-ray detector position : 55mm Working distance : 39mm Accelerating voltage : 20 KV Take-off angle : 27 deg Beam current : 200 picoamps Accumulation time : 100 seconds VWndow : Be A standardless quantitative analysis (SQ) program was used to analyze the X- ray spectra obtained. Since the selenium ratio is always underestimated due to artifact of the program, a correction factor (x186), which has been calibrated using known KIAu/Se ternary compounds, was used to evaluate selenium percentage. The analyses reported are the average of four to six individual measurements on different crystals. 2.3. Synthesis Chemicals were measured and loaded in Pyrex tubes under a. dry nitrogen atmosphere in a Vacuum Atmospheres Dri-Lab glovebox. Potassium 79 monosulfide (K28) and alkali metal monoselenide (Azse; A=Na, K, Cs) were prepared in liquid ammonia from alkali metal and elemental sulfur (or selenium) in a 2:1 ratio. Potassium (1,S-pz-pentasulfldo)aurate(I), KAuSs (I) 0.198 g (1.8 mmol) of K28, 0.196 g (1.0 mmol) of Au powder, and 0.256 g (8.0 mmol) of S powder were mixed together and loaded in a Pyrex tube which was flame- sealed under vacuum («10'3 torr). The tube was placed in a computer- controlled Furnace and heated at 250 0C for 99 hrs and cooled slowly to 50 °C at a rate of 2 OCIhr. The pale yellow needle-like crystals were obtained by removing excess molten potassium polysulfides with water under a N2 atmosphere (yield: 78 % based on the Au used). A quantitative analysis performed on a large number of crystals using scanning electron microscope/EDS gave an average composition of K1_oAu1,oS5,4. Potassium (1,5-p2-pentaselenido)aurate(l), KAuSes (II) 0.157 g (1.0 mmol) of Kass, 0.098 g (0.5 mmol) of Au powder, and 0.316 g (4.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum («10'3 torr). The tube was placed in a computer- controlled Furnace and heated at 250 °C for 99 hrs and cooled slowly to 50 0C at a rate of 2 °CIhr. The reddish black needle-shaped crystals were obtained by removing excess potassium polyselenides with water under a N2 atmosphere (yield: 74% based on the Au used). A quantitative analysis performed on a large number of crystals using scanning electron microscope/EDS gave an average composition of K1.oAu1,oSe5.1. Cesium (1.3112-trlselenldo)aurate(I), CsAuSea (Ill) 0.172 g (0.5 mmol) of 08238, 0.024 g (0.12 mmol) of Au powder, and 0.078 g (1.0 80 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled Furnace and heated at 350 °C for 54 hrs and cooled slowly to 50 0C at a rate of 2 oClhr. The purple colored needle-shaped crystals were obtained by removing excess cesium polyselenides with water under a N2- atmosphere (yield: 63 % based on the Au used). A quantitative analysis performed on a large number of crystals using scanning electron microscope/EDS gave an average composition of Cs1,oAu1,oSe3,2. Tripotassium bls(pentaselenldo)(1,3-p2-trlselenldo)aurlte(lll), K3AuSe13 (N). 0.140 g (0.9 mmol) of K2Se, 0.098 g (0.5 mmol) of Au powder, and 0.316 g (4 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled Furnace and heated at 250 0C for 99 hrs and cooled slowly to 50 °C at a rate of 2 oC/hr. The black needle-like crystals were obtained by removing excess molten potassium polyselenides with water under a N2 atmosphere (yield: 57% based on the Au used). A quantitative analysis performed on a large number of crystals using scanning electron microscope/EDS gave an average composition of lQ,oAu1.1Se12. Potassium bls(p2-selenldo)aurlte(lll), KAuSe2 (V) 0.070 g (0.45 mmol) of K2Se, 0.10 g (0.51 mmol) of Au powder, and 0.158 g (2.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~103 torr). The tube was placed in a computer- controlled Furnace and heated at 290 0C for 99 hrs and cooled slowly to 50 °C at a rate of 2 oClhr. The metallic black needle-shaped crystals were obtained by removing excess potassium polyselenides with water under a N2- atmosphere (yield: 78 % based on the Au used). A quantitative analysis 81 performed on a large number of crystals using scanning electron microscope/EDS gave an average composition of K1,oAu1,oSe2.o. Trisodium bls(trlselenido)(p2-dlselenido)aurlte(lll), Na3AuSea (Vl) 0.031 g (0.25 mmol) of Na2Se, 0.049 g (0.25 mmol) of Au powder, and 0.158 g (2.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~103 torr). The tube was placed in a computer-controlled Furnace and heated at 310 °C for 99 hrs and cooled slowly to 50 0C at a rate of 2 °CIhr. The purple-colored, needle-shaped crystals were obtained by removing excess molten sodium polyselenides with water under a N2 atmosphere (yield: 67 % based on the Au used). A quantitative analysis performed on a large number of crystals using scanning electron microscope/EDS gave an average composition of Na3,oAuo,94Se7,9. Potassium bls(p2-selenldo)aurlte(lll), NaAuSe2 (VII) 0.093 g (0.75 mmol) of Na2Se, 0.197 g (1.0 mmol) of Au powder, and 0.158 g (2.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10:3 torr). The tube was placed in a computer-controlled Furnace and heated at 290 0C for 99 hrs and cooled slowly to 50 0C at a rate of 2 oC/hr. The black rectangular crystals were obtained by removing excess potassium polyselenides with water under a N2 atmosphere (yield: 81 % based on the Au used). A quantitative analysis performed on a large number of crystals using scanning electron microscope/EDS gave an average composition of Na1_oAu1,oSe2,o. 82 2.4. X-ray Crystallographic Studies All compounds were examined by X-ray powder diffraction for the purpose of phase characterization and identification. The d-spacings for each compound were obtained from the powder patterns recorded on a Phillips XRG- 3000 computer-controlled powder diffractometer, operating at 40KV, 35 mA. Graphite monochromated Cu-Ka radiation was used. To verify product homogeneity, the d-spacings observed for the bulk materials were compared, and found to be in accord, with those calculated from the single crystal X-ray structure analysis data. The calculation of d-spacings was performed using the POWDfO program“. The results are summarized in Tables 24-30. 83 Table 24. Calculated and Observed X-ray Powder Diffraction Pattern of KAuS5 H K L amok) 60mm) lllmax (abs) 0 0 2 7.73 7.92 22.7 1 1 0 6.57 6.72 55.1 0 2 0 5.37 5.47 10.6 1 1 2 5.00 5.07 27.3 2 0 0 4.15 4.21 25.4 0 0 4 3.86 3.91 7.4 2 1 1 3.75 3.30 13.7 2 0 2 3.66 3.70 23.9 1 3 0 3.29 3.32 37.6 1 3 2 3.02 3.05 100 2 0 4 2.33 2.35 63.2 0 4 0 2.68 2.71 50.1 0 4 2 2.540 2.557 23.5 1 3 4 2.506 2.526 12.9 2 4 0 2.257 2.273 4.3 3 1 4(3 3 0) 2.204 2.207 13.0 2 4 2 2.167 2.131 17.6 3 3 2 2.109 2.123 23.3 4 0 0 2.077 2.090 22.1 1 3 6 2.029 2.053 13.7 1 5 2 2.011 2.021 16.1 2 4 4(0 0 3) 1.9497 1.9594 17.4 3 3 4 1.9069 1.9070 3.2 0 4 6 1.3610 1.3696 6.2 4 0 4 1.3300 1.3391 4.5 0 6 0 1.7929 1.3024 10.6 2 4 6 1.6934 1.7035 6.3 3 3 6 1.6699 1.6772 14.6 4 4 2 1.6031 1.6146 11.6 84 Table 25. Calculated and Observed X-ray Powder Diffraction Pattern of KAuSe5 H K L 0mm 60mm) lllmax(obs.) 0 0 2 3.21 3.29 11.3 1 1 0 6.35 6.91 34.0 0 2 0 5.64 5.69 13.6 1 1 2 5.26 5.29 39.7 2 1 1 3.91 3.93 33.6 1 2 3 3.57 3.59 10.4 1 3 0 3.44 3.45 13.9 1 3 2 3.13 3.19 32.5 2 0 4 2.97 2.93 100 0 4 0 2.32 2.33 73.7 0 4 2 2.66 2.67 17.9 1 4 1 2.64 2.65 22.5 3 1 4 2.305 2.313 7.5 2 4 2 2.270 2.277 7.7 3 3 2 2.201 2.206 13.9 4 0 0 2.156 2.162 33.6 1 3 6 2.144 15 2 2.111 2.116 11.3 0 0 3 2.053 2 4 4 2.047 2.053 33.6 3 2 5 2.020 2.025 10.3 4 1 3 1.975 3 1 6 1.952 1.956 7.2 0 6 0 1.3321 1.3353 10.3 3 3 6 1.7541 1.7531 12.3 4 4 0 1.7136 1.7132 11.3 1 2 9 1.7023 4 4 2 1.6775 1.6815 7.9 2 6 4 1.5903 1.5925 6.8 1 7 2 1.5569 1.5533 5.2 85 Table 26. Calculated and Observed X-ray Powder Diffraction Pattern of CsAuSea H K L MM) dobs (A) lllmax (obs) 0 2 0 6.39 7.07 9.5 0 2 1 4.94 4.99 29.3 1 1 1 3.76 3.86 33.9 1 3 0 3.64 3.67 30.0 0 4 0 3.44 3.47 73.7 0 4 1 3.09 3.12 67.2 2 0 0 3.00 3.03 63.5 1 3 1 2.97 2.99 100 -2 0 2 2.39 2.91 49.9 1 5 0 2.50 2.51 31.3 1 3 2 2.292 2.307 25.4 -1 3 3 2.229 2.253 25.4 -2 4 2(-2 2 3) 2.214 2.196 39.9 -3 1 2 2.093 2.076 31.3 2 0 2(-3 3 1) 1.955 1.957 32.6 -3 3 2 1.9236 1.9150 23.3 -1 1 4 1.3373 1.3317 22.4 -3 3 3(0 0 4) 1.7771 1.7704 19.0 -3 5 2(0 8 1) 1.6797 1.6821 38.9 86 Table 27. Calculated and Observed X-ray Powder Diffraction Pattern of K3AuSe13 H K L awed) do“ (A) l/lmax (obs) 1 0 0 14.91 16.00 2.6 2 0 0 7.45 7.62 7.6 0 0 2 6.34 6.93 12.9 -1 0 2 6.33 6.52 6.1 -2 0 2 5.22 5.32 4.9 3 o 0 4.97 5.04 21.7 1 1 0 4.61 4.63 13.4 0 1 1 4.57 4.64 17.3 -1 1 1 4.40 4.46 32.5 1 1 1 4.34 4.40 17.3 2 1 0 4.07 4.12 14.9 0 1 2 3.96 4.00 10.6 -1 1 2(2 1 1) 3.36 3.90 32.2 -2 1 2 3.55 3.59 6.3 3 1 0 3.47 3.51 17.0 2 1 2 3.44 3.43 21.7 0 0 4 3.42 3.44 31.0 -1 1 3 3.23 3.31 6.5 1 1 3 3.21 3.24 7.9 -3 1 2 3.15 3.13 100 -2 1 3 3.09 3.12 20.3 2 0 4 3.03 3.06 24.0 4 1 1 2.92 2.94 31.7 3 0 4 2.73 2.75 45.0 3 1 3 2.70 2.72 22.4 -3 1 4 2.50 2.52 6.6 1 2 0 2.397 2.413 11.2 1 2 1 2.357 2.373 19.6 87 Table 27. (cont'd) H K L doaHA) dobs (A) lllmax(obs.) 5 1 2 2.339 2.353 13.3 2 2 0 2.309 2.323 6.2 -2 2 1 2.235 2.299 13.7 1 0 6 2.233 2.247 5.3 3 2 0 2.132 2.196 22.0 -6 1 2 2.143 2.156 13.3 -3 2 2 2.097 2.109 15.6 -6 0 4 2.073 2.091 27.4 6 1 2(0 1 6) 2.069 2.079 45.0 -3 2 3 1.991 2.037 13.4 5 1 4 1.937 1.997 9.4 4 1 5 1.953 1.963 3.1 -3 1 6 1.9493 1.957 9.3 -4 2 3 1.3347 1.3943 12.4 -5 2 1 1.3767 1.3743 13.6 0 2 5 1.3170 1.3217 13.3 1 1 7 1.7377 1.7953 7.2 3 0 2 1.7690 1.7771 5.0 7 1 3 1.7555 1.7636 10.4 -3 1 7 1.7397 1.7475 9.5 4 2 4 1.7213 1.7301 11.7 -6 2 2 1.7031 1.7109 14.3 622 1%% 16N5 w] 9 0 0 1.6572 1.6635 30.7 -3 0 3 1.6523 1.6533 26.7 4 2 5 1.6055 1.6091 9.5 804 1 .5929 1.5996 12.6 88 Table 28. Calculated and Observed X-ray Powder Diffraction Pattern of KAUSez H K L MGM) dob,3 (A) lllmax (obs) 1 1 0 5.44 5.50 100 2 0 0 3.34 3.33 56.3 2 1 0 3.44 3.46 42.3 1 11 3.04 3.05 10.4 2 2 0 2.72 2.73 55.5 2 0 1 2.65 2.66 69.5 2 1 1 2.50 2.52 21.3 3 1 0 2.434 2.447 5.9 2 2 1 2.135 2.194 26.1 3 1 1 2.027 2.037 74.0 4 0 0 1.9247 1.9323 4.3 4 1 0 1.3672 1.3731 4.5 0 0 2 1.3325 1.3391 10.6 3 3 0 1.3146 1.3217 40.3 1 1 2 1.7367 1.7422 10.4 4 2 0 1.7215 1.7276 36.2 4 0 1 1.7040 1.7101 33.3 2 0 2 1.6545 1.6534 6.0 2 1 2 1.6176 1.6222 3.3 4 2 1 1.5582 1.5638 13.7 89 Table 29. Calculated and Observed X-ray Powder Diffraction Pattern of NaaAuSea H K L 0mm) dobs (A) l/lmax(obs.) -1 1 1 7.32 7.46 39.4 -2 0 2 6.96 7.03 19.4 2 0 0 5.06 5.12 11.5 -3 1 2 4.72 4.73 47.5 1 1 1 4.19 4.24 14.6 3 1 3 4.03 4.12 33.5 -4 0 2 4.03 4.07 13.4 -2 2 2 3.66 3.69 25.1 -1 1 3 3.22 3.25 16.6 3 1 0 3.14 3.16 19.7 -3 1 4 3.10 3.12 10.3 -4 2 3 2.99 3.02 25.2 -4 2 2 2.94 2.96 100 -2 0 4 2.91 2.93 45.5 1 3 0 2.76 2.77 15.7 -4 2 4 2.70 2.72 17.3 -5 1 5 2.64 2.66 17.2 2 2 1 2.62 2.64 7.3 1 3 1 2.464 2.473 14.6 -3 3 1 2.450 2.457 10.3 3 1 1 2.405 2.419 10.5 -6 2 4 2.361 2.374 24.1 -7 1 5 2.317 2.329 7.3 -7 1 4 2.296 2.307 12.0 -6 2 3 2.279 2.291 5.2 -1 3 3 2.213 2.225 12.5 1 1 3 2.130 2.137 3.3 0 4 0 2.152 2.163 23.7 - 90 Table 29. (cont'd) H K L dcab(A) dam (A) lllmax(obs.) -4 0 6 2.136 2.147 23.1 2 2 2 2.099 2.109 30.1 0 4 1 2.033 2.092 30.1 -6 2 6 2.042 2.052 29.3 -3 0 4 2.015 2.025 14.9 -5 3 5 1.993 2.007 19.3 2 4 0 1.930 1.933 9.4 -7 1 7 1.9379 1.9453 11.7 -4 2 6 1.9139 1.9225 21.0 -5 3 1 1.3745 1.3319 25.2 -7 3 5 1.3433 1.3513 22.3 -3 2 4 1.3251 1.3344 20.6 -5 3 6 1.7345 1.7920 16.9 -7 3 6 1.7635 1.7753 13.1 -3 2 7 1.7633 1.7471 19.7 -3 0 3 1.7404 1.7129 3.1 -10 0 6 1.6326 1.6904 16.6 5 3 0 1.6544 1.6591 16.0 4 0 2 1.6525 1.6546 15.7 1 5 1 1.6212 1.6269 24.1 -3 5 3(-3 2 3) 1.6144 1.6196 59.7 -6 4 2 1.5333 1.5931 12.5 -15 3(-9 3 7) 1.5433 1 .5479 23.7 91 Table 30. Calculated and Observed X-ray Powder Diffraction Pattern of NaAuSe2 H K L amuA) dobs (A) l/lmax(obs.) 1 0 o 6.74 6.36 4.3 1 1 0 5.25 5.32 53.6 01 1 5.12 5.19 30.7 1 1 1 3.72 3.76 6.5 1 2 0 355 3.53 9.3 2 0 0 337 3.40 9.3 -1 2 1 331 3.34 7.1 -2 1 1 313 3.15 6.8 1 2 1 295 2.97 100 -2 0 2 272 2.73 374 -2 1 2 253 2.60 272 1 1 2 2537 2.55 77 -3 1 1 2.237 2.250 137 -3 0 2 2.123 2.140 116 0 1 3 2.093 2.104 153 0 4 0 2.093 2.097 314 2 0 2 2.030 2.039 24 3 -3 1 2 2.062 2.074 13 5 -3 2 1 2.030 2.041 31 1 0 4 1 1.992 2.001 5 9 -1 2 3 1.973 1.934 27 6 2 3 1 1.9461 1.956 163 1 3 2 1.9267 1.9370 73 -3 3 1 1.7349 1.7923 137 3 2 1 1.7323 1.7364 190 -1 3 3 1.7457 1.7530 3 2 0 3 3 1.7095 1.7164 6.5 4 0 2 1.6333 1.6954 12.9 -2 3 3 1.6712 1.6765 7.3 -2 4 2 1.6595 1.6613 27.9 1 5 0 1.6254 1.6314 20.6 4 2 1 1.6123 1.6196 17.0 92 The X-ray single crystal data of KAuSs, KAuSes, K3AuSe13, Na3AuSea, NaAuSez, and KAuSe2 were collected on a Nicolet P3 four circle diffractometer with graphite monochromated Mo-KCl radiation using the 0-20 scan mode. The data for CsAuSea were collected on a Rigaku AFC68 diffractometer with graphite monochromated Mo-Ka radiation using the 0-26 scan mode. Accurate unit cell parameters for all compounds were obtained from the least-squares refinement of the 26, m, )0 and 0 values of 20-25 machine-centered reflections. The stability of the experimental setup and crystal integrity were monitored by measuring three standard reflections periodically (every 100 reflections) during the data collection period. The intensities did not show any appreciable decay. An empirical absorption correction based on 141 scans for 3 reflections was applied to the KAuSs data. Two absorption corrections were applied to the data of KAuSes, lQAuSe13, Na3AuSe3, NaAuSe2, KAuSe2, and CsAuSea: an empirical absorption correction based on 111 scans for 3 reflections followed by a DIFABS57 correction. The structures of KAuSs, KAuSe5, KunSeta, NaaAuSee, NaAuSe2, and KAuSe2 were solved with direct methods using SHELXS-8658 and were refined with the SDP59 package of crystallographic programs. Computations were performed on a VAXstation 2000 computer. The structure of CsAuSea was solved with direct methods using SHELXS-86 and was refined with the TEXSAN60 package of crystallographic programs on a VAXstation 3100 computer. All atoms were refined anisotropically. The complete data collection parameters and details of the structure solution and refinement for all compounds are given in Table 31. The final coordinates, temperature factors, and their estimated standard deviations of all atoms are shown in Tables 32-37. 93 Table 31. Summary of Crystallographic Data for KAuSs, KAuSes, CsAuSea, KgAuSe13. KAuSe2, NaaAuSeg, and NaAuSe2 commnd I II III Formula KAuSs KAuSe5 CsAuSea Formula weight 396.39 630.87 566.75 space group lbam lbam C2/c a (A) 3.310(2) 3.625(6) 6.433(3) b (A) 10.753(3) 11.293(9) 13.739(5) c(A) 15.463(4) 16.425(11) 7.651(3) 01 (deg) 90.0 90.0 90.0 8 (deg) 90.0 90.0 1 1217(3) 7 (deg) 90.0 90.0 90.0 Vol (A3), 2 1332.7(6), 3 1600(2), 3 633.3(5), 4 Temperature (°C) 23 23 23 Crystal size (mm) 0.44x0.16x0.08 0.30x0.1 6x0. 14 0.49x0.03x0. 05 Radiation Mo-Ka Mo-Ka Mo-Ka u (Mo-K0, cm'1) 231.8 413.1 457.2 Dcalc (gch 3.81 5.24 5.94 20",” (deg) 50 46 50 Scan method 0/26 6/20 0/26 No. of data collected 2829 744 121 1 No. of unique data 639 536 581 No. of data used 519 382 390 (F02>30(F02)) No. of atoms 6 6 4 No. of variables 37 37 25 Phasing technique Direct methods Direct methods Direct methods Final RlRw 5.1/6.3 5.7/7.1 4.0/4.7 Max. shift/esd 0.0 0.0 0.00 (final cycle) ' Extinction coefficient 4.22x10'7 1.61x10'7 NIA Table 31. (cont'd) 94 comthnd IV V VI Formula K3AuSe13 KAuSe2 NaAuSeg Formula weight 1340.75 393.99 897.62 space group P2/c P4/m bm C2/c a (A) 14.949(6) 7.699(3) 16. 943(3) 0 (A) 4.3539(2) 7.699(3) 8.610(3) c (A) 13.723(5) 3.665(1) 13.931(5) (1 (deg) 90.0 90.0 90.0 8 (deg) 9886(3) 90.0 143.32(2) 7 (deg) 90.0 90.0 90.0 Vol (A3), 2 994.3(7), 2 217.2(2), 2 1214.3(3), 4 Temperature (°C) 23 23 23 Crystal size (mm) 0.42x0.10x0.07 0.18x0.18x0.05 0.38x0.20x0.04 Radiation Mo-Ka Mo-Ka Mo-Ka (1(Mo-Ka, cm'1) 315.6 511.4 359.0 Dcalc @1ch 4.48 6.02 4.91 20",” (deg) 48 48 46 Scan method 8/20 6/26 8/26 No. of data collected 1918 460 1924 No. of unique data 1570 1 18 858 No. of data used 1167 111 698 (F02>30(F02)) No. of atoms 10 3 7 No. of variables 81 1 0 59 Phasing technique Direct methods Direct methods Direct methods Final R/Rw 4.4I5.0 3.2/3.7 5.9/6.8 Max. shift/esd 0.0 0.00 0.0 (final cycle) Extinction coefficient 2.72x10'7 1.00x10'7 5.22x10'7 - 95 Table 31. (cont'd) cornmnd Vii Formula NaAuSe2 Formula weight 377.88 space group P21/c a(A) 6.991(3) b(A) 3.374(9) c(A) 6.724(3) 01 (deg) 90.0 8 (deg) 105.23(9) y (deg) 90.0 Vol (A3), 2 379.3(7), 4 Temperature (0C) 23 Crystal size (mm) 0.20X0.16X0.08 Radiation Mo-Ka p. (Mo-Ka, cm'l) 575.1 Dcalc (glcm3) 6.61 29rnax (deg) 43 Scan method 8/26 No. of data collected 1430 No. of unique data 604 No. of data used 510 (F02>30(F02)) No. of atoms 4 No. of variables 38 Phasing technique Direct methods Flnal R/Rw , 6.8/8.3 Max. shift/esd 0.00 (last cycle) Extinction coefficient 4.26 x107 96 Table 32. Fractional Atomic Coordinates and Beq Values for KAuS5 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Au 0 0.13773(9) 1/4 174(2) S(1) 0.2012(7) 0.1402(5) 0.3517(4) 2.1(1) 5(2) 0.2943(6) 0.1760(5) 0.1067(3) 137(9) S(3) 0.9422(9) 0.3331(7) o 2.1(1) K(1) 1/2 0 1/4 23(2) K(2) 0 0 0 2.5(2) 8' B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[aZB11 + £32822 + 02833 + ab(cos 10812 + aclcos 0813 + bc(cos (1)323]. Table 33. Fractional Atomic Coordinates and Bag Values for KAuSes with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 AU 0 0.1306(2) 1/4 239(4) 38(1) 0.2065(5) 0.1286(4) 0.3488(3) 255(8) Se(2) 0.2337(5) 0.1701(4) 0.1125(3) 225(3) 88(3) 0.9500(6) 0.3418(6) 0 2.5(1) K(1) 1/2 0 1/4 2.4(4) K(2) 0 0 0 3.1(4) a B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[aZB11 + sz22 + 02833 + ab(cos 10812 + ac(cos 0813 + bc(cos 008231. 97 Table 34. Fractional Atomic Coordinates and ng Values for CsAuSe3 with Their Estimated Standard Deviations in Parentheses Atom x _y z Baqa, A2 AU 0 0.5780(1) 1/4 2. 27(5) 36(1 ) 0.21 60(4) 0.5887(2) 0.0557(3) 2. 7(1 ) 36(2) 0.5 0. 6950(2) 1/4 2. 6(1 ) Cs -0.5 0.6476(2) -1/4 334(9) a B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Baq = (4/3)[aZB11 + b2822 + 02833 + ab(cos 10812 + £1de B)B13 + bc(cos @823]- 98 Table 35. Fractional Atomic Coordinates and Baq Values for K3Au(Se5)2(Se3) with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 AU “2 0 0 1 .71 (2) 38(1) 0.6198(1 ) 0.3099(4) -0.0482(1 ) 215(4) 88(2) 0.7325(1 ) 0.3674(5) 0.0779(2) 2. 53(4) 38(3) 0.1 505(2) 0.0742(5) 0.4666(2) 2. 99(5) 38(4) 0.0730(1) 0.3259(5) 0.5603(2) 313(5) 38(5) 0.1 522( 1 ) 0.3037(5) 0.7334(2) 2. 56(4) 38(6) 0.4140(1) 0.8955(5) 0.3431(1) 247(4) Se(7) 1/2 0.3971(6) 0.750 231(5) K(1) 0 0.172(1) 1/4 3.2(1) K(2) 0.7130(3) 0.203(1) 0.3346(4) 3.2(1) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4l3)[a2811 + b2822 + 02333 + ab(cos 10312 + 30(008 £9313 + b0(COS (1)323]- 99 Table 36. Fractional Atomic Coordinates and Baq Values for KAuSe2 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 AU "2 0 0 0.20(1 ) Se 0.347 0.1523(3) 1/2 049(3) K 0 0 0 1 2(1) a B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[a2811 + sz22 + 02833 + ab(cos 10812 + ac(cos B)B1a + bclcos 00823]. Table 37. Fractional Atomic Coordinates and Beq Values for NaaAuSeg with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Au 1/4 1/4 1/2 072(3) 38(1) 0.4086(2) 0.4569(3) 0.6091(2) 090(6) Se(2) 0.1014(2) 0.3226(4) 0.2190(2) 117(6) 38(3) 0.1966(2) 0.4691(4) 0.7269(2) 107(6) 88(4) 0.3802(2) 0.7542(4) 0.7587(2) 1 .51 (6) Na(1) 1/2 1/2 1/2 2.9(4) Na(2) 0.1358(9) 0.602(2) 0.458(1) 3.1(4) 8* 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[a2811 + b21322 + 02833 + ab(cos 10812 + ac(cos B)B1a + bc(cos 00323]. 100 Table 38. Fractional Atomic Coordinates and Beq Values for NaAuSe2 with Their Estimated Standard Deviations in Parentheses Atom x y z nga, A2 AU 0.2176(2) 0.631 1(1) 0.0087(2) 040(2) Se(1) -0.0205(5) 0.5764(4) 0.2122(5) 036(6) 38(2) 0.4157(5) 0.8314(4) 0.2506(5) 0. 72(6) Na 0.275(2) 0.366(2) 0.488(2) 1 .9(3) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4I3)[aZB11 + b2822 + 92833 + ab(cos 10812 + ac(oos £0813 + bc(oos 008231. 101 3. Results and Dlscusslon 3.1. Synthesis and Spectroscopy The synthesis of all compounds has been accomplished at the intermediate temperature range (from 250 0C to 350 00) using alkali metal polychalcogenide fluxes as solvents and reagents as shown in eq 1. M20 + mAu + qQ ------> AaAqu eq 1. (A=Na, K, Cs; Q=S, Se) Each reaction formed a complete melt at the employed temperatures (250 °C~350 °C). The equation given above is not balanced and shows only reactants used (left-hand side) and product obtained (right-hand side). Since alkali metal polychalcogenide flux is very reactive and corrosive, it oxidizes Au metal while itself is reduced to shorter polychalcogenides. After reaction, excess alkali metal polychalcogenide flux is easily removed from the product by washing with water and/or dimethylformamide (DMF). All compounds are stable in the air and moisture for several days. In this study we have seen profound redox chemistry associated with Au+’3+ ions and controlled by the chemical composition of alkali metal polychalcogenide fluxes. When we tried to optimize the reaction conditions for the Au+ compound of KAuSe5, which was prepared from the reactant ratio KZSe/Au/Se of 2/1/8 at 250 °C, a slight variation of K286 ratio from 2 to 1.8 yielded a surprising result: the novel Au3+ compound of K3AuSe13. The subsequent isolation and structural characterization of three more Au3+ compounds in KIAu/Se and Na/Au/Se systems and another Au+ compound in 102 the Cs/Au/Se system suggests that a variety of Am and Au3+ compounds may occur in these molten polychalcogenide fluxes which may undergo complex solution equilibria. Various alkali metal polychalcogenide flux compositions AzQx (x=3~9) are obtained by fusing A20 with stoichiometric amounts of chalcogenide atoms as shown in eq 2. nAzQ + 80 -------> nAzQx (n=1~4; x=9~3) eq 2. The reactivity of these fluxes varies according to their chemical compositions. Polychalcogenide fluxes with longer chains, AzQx (x=9, 5), generally favored compounds with long polychalcogenide ligands (sz' x=5,3). For example, KAuSe5, KaAuSe13, and NaaAuSeg were prepared from K2385, K28e5,4 and N22369 fluxes, respectively. The opposite is also true. Shorter chain fluxes generally favored monochalcogenide compound formation. For example, NaAuSe2 was prepared from Na2863.7. but KAuSez was prepared from the longer polychalcogenide flux (K28e5,4), the composition used for preparation of K3AuSe13. However, the increased amounts of Au required for preparation of KAuSe2 broke the starting K2865,4 flux into a much shorter chain polychalcogenide flux during the reaction. Another interesting observation made during this study is the alkali metal cation size effect which could support new structural frameworks (e.g. one-dimensional KAuSe2 vs layered NaAuSe2) and various polychalcogenide ligands (e.g. 8e32' vs Sesz'). Therefore, the systematic investigations with various size of alkali metals turned out to be quite valuable. 103 In the far-IR region all compounds exhibit spectral absorptions due to Q-Q and Au-Q stretching vibrations. The observed absorptions are given in Table 39 and their spectra are shown in Figures 1820. Table 39. Frequencies (cm") of Spectral Absorptions of (I), (II), (III), (IV), (V), (VI), and (VII) due to Q—Q and M-Q stretching Vibrations (I) ('0 (III) (IV) (V) (Vl) (W) KAuSs KAuSe5 CsAuSea K3AuSe13 Na3Au8e3 KAuSe2 NaAuSe2 461 (s)8 257 (s) 243 (s) 264 (s) 259(3) 231 (s) 236 (s) 317(m) 244 (sh) 233 (m) 231 (s) 224 (m) 146 (m) 230 (m) 255 (m) 225(s) 207(w) 222 (w) 210(m) 110(m) 222 (sh) 212 (m) 213 (w) 163 (w) 214 (w) 194 (w) 191 (m) 194 (w) 177(m) 134 (m) 111(m) 8‘ (s) strong; (m) medium; (w) weak; (sh) shoulder The only polysulfide compound in this study, KAuSs, shows an absorption at 461 cm'1 which is attributed to a SS stretching vibration by comparison with the spectra of other known polysulfide compounds.64 (see also Table 17) The v3-3 vibrational frequencies are observed in the range of 446~500 cm'1. The additional peak at 317 cm'1 can be attributed to an Au-S stretching vibration. For the Au polyselenide compounds, a spectral absorption is observed in the range of 242~264 cm'1. This band can be assigned to a Se-Se stretching 104 75 (A) .. fl“) 2 '- 59 4 5‘ T l l 550 462 374 236 193 tan") w A (C) 2'. .- l l l 350 304 253 212 166 1cm") Figure 18. Far-IR spectra of (A) KAuSs, (B) CsAuSea, and (C) KAuSes TRANSMITI'ANCE (96) 105 (A) (B) I l l 350 303 256 209 162 WAVENUMBER (cm") Figure 19. Far-IR spectra of (A) KaAuSe13 and (B) NaaAuSee TRANSMITTANCE (96) 106 (A) (B) l I l 350 288 226 164 102 WAVENUMBER (cm") Figure 20. Far-IR spectra of (A) KAuSe2 and (B) NaAu862 107 vibration by comparison with the spectra of other known polyselenide complexes and with that of the unbound ligand (Ph4P)28e5 (1739-39 267 cm'1) 26M. The 1739-39 absorption in this region has been observed previously in various compounds, eg. Sex2'65 (x=1-6) at 258 cm'l, c-Se666 at 253 cm'1, [Fezse12]2'24 at 258 cm'1, [Pd(Se4)2]2'67 at 274 cm'1, and a-[CuSe4lnn’ at 246~252 on1 (see Table 17). In addition, an absorption in the vicinity of 230 cm'1 found in all Au/Sexz‘ compounds may be attributed to an Au-Se stretching vibration. The monoselenide compounds of (V) and (VII) show common absorptions around 230 cm'1 and 110 cm”. Using simple group theory and assuming D4). symmetry on the Au3+ center, we expect one IR-active Au-Se stretching vibration in KAuSe2 . For the NaAuSe2 compound there are 4 IR- active Au-Se stretching vibrations expected by assuming the C211 symmetry on the Au3+ center. Thus, we can assign the 230 cm'1 band in KAuSe2 and the 236 cm'1, 230 cm'1,and 222 cm'1 bands in NaAuSe2 as Au-Se stretching vibrations. This interpretation must be an oversimplication on this complicated extended lattice system. It is usually difficult to interpret the IR spectra of metal polychalcogenide compounds without ambiguity. This is because of the fact that M-Se and Se-Se stretching frequencies fall in the same low-frequency IR region (200-340 cm”) and that systematic IR spectroscopic data for the various free ligands (0x2: x=2-6) and metal chalcogenide complexes are still lacking. 3.2. Description of Structures 3.2.1. Structure of KAuSs (I) and KAuSes (II) (I) and (II) are isostructural. They possess one-dimensional polymeric chains. Figure 21 shows the alternating planes of charge compensating K+ ions and [AuQslnn' chains along the crystallographic b-axis in the unit cell. 108 Figure 21 . ORTEP representation of the unit cell of KAuQs (st, Se). 109 The anionic [AuQslnn' chains are composed of pentachalcogenide ligands (852' and 3852' respectively) bridging Au+ atoms via their terminal 0 atoms. The chains lack a center of symmetry and run parallel to the crystallographic c- axis. The Au+ centers, situated on a crystallographic 2-fold axis, show a linear coordination with a Q-Au-Q bond angles of 178.7(2) deg and 178.9(2) deg for (I) and (II), respectively. The observed Au-Q bond distances (Au-S; 2.296(6) A and Au-Se; 2.410(4) A) are similar to those found in KAuQ77 (Q=S, Se) at 2.305(6) and 2.414(1) A, respectively. There are short Au-Au contacts at 2.963(1)A and 2.950(3)A for (I) and (II) respectively. These contacts are considerably shorter than those observed in KAuS77 (3.260 (1) A) and a-AuSe83 (3.222 (6) A) and comparable to that in AggAuSez (2.974 .5068“). These close Au-Au interactions occur between chains along the crystallographic b-axis, to form dimers as shown in figure 22. These close Au-Au contacts are considered as favorable d10-d10 interactions68. There are no interactions between the chains along the crystallographic a-axis. Selected bond distances and angles for (I) and (II) are given in Table 40. The two crystallographically distinct K atoms in the asymmetric unit participate in ionic interactions with the 052' ligands. The K atoms are surrounded by eight chalcogen atoms. The shortest K-S distances in (l) are K(1)-S(1) 3.304(5) A, and K(2)-S(1) 3.214(5) A; the shortest K-Se distances in (II) are K(1)-Se(1) 3.339(4) A, and K(2)-Se(1) 3.334(4) A. 110 .coszEoEfi .o moon. on. @5266 .265 2.21.03: .0 222:6 .mcoacoEfiocO .mm 05mm 85 )--——— " 111 Table 40. Selected Bond Distances (A) and Angles (deg) in KAuSs and KAuSes with Standard Deviations in Parentheses KAuSs Au-S(1) 2.296(6) S(1)-Au-S(1) 173.7(2) S(1 )-S(2) 2.080(7) Au-S(1)-S(2) 103.7(3) S(2)-S(3) 2.063(7) S(1)-S(2)-S(3) 107.9(3) S-S (mean) 2.07 S(2)-S(3)-S(2) 106.3(4) Au-Au 2.963(1) K(1)-S(1) 3.305(5) (x4) K(2)-S(1) 3.214(5) (x4) K(1)-8(2) 3.379(5) (x4) K(2)-8(2) 3.506(5) (x4) K(1)-S (mean) 334(4) K(2)-S (mean) 3.36 KAuSes Au-Se( 1) 2.410(4) Se(1)-Au-Se(1) 179.0(2) Se(1)-Se(2) 2.363(6) Au-Se(1)-Se(2) 101 .4(2) Se(2)-Se(3) 2.343(5) Se(1)-Se(2)-Se(3) 104.2(2) Se-Se (mean) 235(1) Se(2)-Se(3)-Se(2) 104.2(3) Au-Au 2.950(3) K(1)-Se(1) 3.339(4) (x4) K(2)-Se(1) 3.334(4) (x4) K(1)-Se(2) 3.503(4) (x4) K(2)-Se(2) 3.619(4)(x4) K(1)-Se (mean) 342(9) K(2)-Se (mean) 3.50(13) 112 3.2.2. Structure of CsAuSes(lll) The anion of this compound also has a one-dimensional polymeric structure, as shown in Figure 23. Like the homologous [Au(Se5)]n"' in (II), the anionic [Au(Se3)]n"' chains are composed of Se32' ligands bridging adjacent Au atoms via terminal Se atoms. The chains lack a center of symmetry and run parallel to the crystallographic a-axis in a helical fashion. The Au atoms, situated on a crystallographic 2-fold axis, have a slightly distorted linear geometry with a Se-Au-Se bond angle of 172.9 (1) deg. Since there are no close Au--Au or interchain Au--Se contacts observed in this compound, this distortion may be due to close Au-Cs contacts which are in the range of 3.784(3) ~ 4.087(2) A. The sum of their van der Waals radii is approximately 4.99 A (The van der Waals radius of Cs was estimated using the linear correlation between known ionic and van der Waals radii of alkali metals). The Au-Se distance is 2.401(2) A which is similar to that found in (II). Selected bond distances and angles are given in Table 41. There is one crystallographically distinct Cs atom in the asymmetric unit. The Cs atoms participate in ionic interactions with the Se32' ligands. Within the limit of 3.9 A, the Cs atom has eight Se atoms around it. The shortest Cs-Se distance in (III) is with terminal Se atoms (Cs-Se(1)) at 3.573(2) A. 113 528% = 5. e38 ooze; av 52696 80: £506 8303 A3 59wng B 9503 .958 92 6 8sz08.59 dmhmo .mm 059... E. .5 114 Table 41. Selected Bond Distances (A) and Angles (deg) in CsAuSe3 with Standard Deviations in Parentheses Au-Se(1) 2.401(2) (x2) Se(1)-Au-Se(1) 172.9(1) Au-Cs 3.784(3) Au(1)-Se(1)-Se(2) 98.4(1) Se(1)-Se(2) 2.384(3) (x2) Se(1)-Se(2)-Se(1) 104.1(2) Cs-Se( 1) 3.573(2) (x2) Cs-Se(2) 3.881(2) (x2) Cs-Se(1) 3.763(3) (x2) Cs-Se(2) 3.900(2) (x2) Cs-Se (mean) 3.78(14) 115 3.2.3. Structure of KgAuSeu (IV) The compound of K3AuSe13 is very intriguing and the most Se rich Au compound yet characterized to the best of our knowledge. The structure is shown in Figure 24. This is a Au3+ compound composed of centrosymmetric one-dimensional [Au(Se3)(Ses)2]n3"' chains and charge balancing K+ ions. The chains are composed of Se32' ligands bridging adjacent Au atoms to which two 8e52' ligands are coordinated in a trans fashion via one of their terminal Se atoms. The Au atoms are situated on a crystallographic inversion center. The chains run parallel to the crystallographic c-axis. Two views of one-dimensional [Au(Se3)(Se5)2]n3n' chains are shown in Figure 25. The geometry around Au atoms is square planar as expected for Au3+. No short Au-Au interactions are observed in the structure. The average Au-Se distance of 248(1) A is comparable to the corresponding distances in ot-AuSe83 (2.48A) and in the molecular [Au23e1012-. 23 (2.45A), both of which are formally Au3+ compounds. Selected bond distances and angles are given in Table 42. The unusual dangling monodentate 8852’ ligands are stabilized by ionic interactions with two crystallographically distinct K+ ions. Figure 26 shows the coordination environments of the K atoms. K(1) atoms are surrounded by 10 Se atoms of which four sets of Se(4)-Se(5) bonds of monodentate 3852’ ligands belong to four different chains. The shortest K-Se interactions are those with unbound Se(5) atoms at 3.262(6) and 3.432(6) A. The K(2) atoms are located in the pockets created by two monodentate Se52' ligands which belong to two different chains. The shortest interactions are with unbound terminal Se(5) atoms at 3.261(6) A and 3.306(6) A. 116 909:3. .o :8 ES oz. .0 2630905 em 2sz . o t50'lu o -..u‘.~lr.m' 9, 1...-.. ...... 117 30(4) e . o o. o O . O 9' . “a. "W . . . . 80m. . . . . . . . - . hut . . , . “' . 4- . o o o O O O m . . . . . ° '10) Se(1) 5°“) 5°“) 8 9‘3 We ’ ”-1 7 ‘ ‘3 50(4) iv 3 ‘ ‘~b‘\ /o 7” V 3 0 «EV. 04<§§a Se(5) .13) 0 ' 05/ Na. ’ g " ‘lakxv e" 9!". .La/o;>""O—. Figure 25. ORTEP representation of two views of the one-dimensional [AuSe1aln3"' chain. 118 .oEocom 9:36. 2:3 mEQm v. .o mEoEco._>co 5:26.000 .8 050E Eom £8 Sow Eom 4.. £3 . 0v vu. a» 56m 86m ..w 58 Ex H Aw 56m Ex . A 6 v , «a. n, 52.... AW 1 04. £8 a 53 Ed 119 Table 42. Selected Bond Distances (A) and Angles (deg) in KunSe13 with Standard Deviations in Parentheses Au-Se(1 ) 2.465(2) (x2) Se(1)-Au-Se(1 ) 180.0(0) Au-Se(6) 2.485(2) (x2) Se(1)-Au-Se(6) 90.8(1 ) Au-Se (mean) 2.48(1 ) Se(6)-Au-Se(6) 180.0(0) Se(1)-Au-Se(6) 89.2(1) Se(1)-Se(2) 2.349(3) Au-Se(1)-Se(2) 1 1 1 .9(1) Se(2)-Se(3) 2.368(3) Se(1)-Se(2)-Se(3) 104.1(1) Se(3)-Se(4) 2.348(3) Se(2)-Se(3)-Se(4) 105.3(1) Se(4)-Se(5) 2.343(3) Se(3)-Se(4)-Se(5) 109.2(1) Se-Se (mean) 235(1) Au-Se(6)-Se(7) 108.9(1) Se(6)-Se(7)-Se(6) 105.6(1) Se(6)-Se(7) 2.351(3) (x2) K(2)-Se(1 ) 3.356(6) K(2)-Se(1 ) 3.266(6) K(1)-Se(3) 3.637(2) (x2) K(2)-Se(2) 3.632(6) K(1)-Se(4) 3.579(5) (x2) l<(2)-Se(3) 3.521(5) K(1)-86(4) 3.592(5) (x2) K(2)-Se(5) 3.306(6) K(1)-Se(5) 3.262(6) (x2) K(2)-Se(5) 3.261 (6) K(1)-Se(5) 3.432(6) (x2) K(2)-Se(6) 3.373(5) K(1)-Se (mean) 3.50(15) K(2)-Se (mean) 3.38(15) 120 3.2.4. Structure of KAuSe2 (V) The structure of KAuSe2 is one-dimensional and contains only monoselenide ligands. It is composed of centrosymmetric one-dimensional [AuSezlnn' chains and charge balancing K+ ions as shown in Figure 27. The one-dimensional [AuSezlnn' chain has a PdClg-type84 structure and is isostructural to [PtSzlnzn' 85 chain. The flat ribbons of [AuSezlnn' are made of edge-sharing AuSe4 square planes. The Au3+ center, situated on a crystallographic mmm site, is strictly square planar as expected for a d8 configuration. The Au-Au distance in the [AuSezlnn' chain is 3.665(1) A, indicating no Au-Au interaction. The Au-Se distance at 2.475(1) A is similar to that in (IV). Selected bond distances and angles are given in Table 43. The charge balancing K+ ions participate in ionic interactions with Se atoms at 3.448(1) A. The geometry around the K atoms is square prismatic. Table 43. Selected Bond Distances (A) and Angles (deg) in KAuSe2 with Standard Deviations in Parentheses Au-Se 2.476(1) (x4) Se-Au-Se 180.0(0) (x2) Se-Au-Se 8448(4) (x2) Se-Au-Se 9552(4) (x2) Au-Se-Au 9552(5) K-Se 3.448(2) (x8) 121 Figure 27. ORTEP representation of the unit cell of KAuSe2. 122 3.2.5. Structure of NaaAuSea (VI) This compound is homologous to K3AU3813. It is composed of centrosymmetric one-dimensional [AU(S€2)(883)2]n3"' chains and charge balancing Nat ions. Two views of one-dimensional [Au(Se2)(Se3)2]n3"' chains are shown in Figure 28. The chains are composed of Se22' ligands bridging adjacent Au atoms to which two 3832' ligands are coordinated in a trans fashion via one of their terminal Se atoms. The Au atoms are sitting on a crystallographic inversion center. Like [Au(Se3)(Se5)2]n3"' in (IV), [Au(Sez)(Se3)2jn3n' represents another remarkable example of a monodentate long polyselenide (Se32‘), where one terminal atom is coordinated to a metal while the other is loose, ionically interacting with alkali ions. The geometry around Au3+ centers is square planar as expected for a d8 configuration. The average Au-Se distance is 249(1) A, similar to those found in Au3+ compounds of (IV) and (V). Selected bond distances and angles are given in Table 44. The coordination environments of two crystallographically distinct Na atoms are shown in Figure 29. Na(1) has distorted octahedral geometry with ionic interactions with selenium atoms in the range of 2.937(4)-3.099(2)A. Na(2) has distorted tetrahedral geometry with ionic interactions with four selenium atoms in the range of 2.94(1)-3.20(2)A. 123 86(4) ~ 86(3) Se(2) Se(1) ' Au - Se(3) - Se(2) - . . \1; 1% . ' ' Se(4) l'> ‘l e(1) 0 ° ., . /v U o . -~ ‘ '\~ - .- Figure 28. ORTEP representation of two views of the one-dimensional [AuSeajn3'i‘ chain. 124 Table 44. Selected Bond Distances (A) and Angles (deg) in Na3AuSe3 with Standard Deviations in Parentheses Au-Se(1 ) Au-Se(2) Au-Se (mean) Se(1)-Se(1) Se(2)-Se(3) Se(3)-Se(4) Se-Se (mean) Na(1 )—Se(1 ) Na(1 )-Se(3) Na(1 )-Se(4) 2.481(3) (x2) 2.499(3) (X2) 2.49(1 ) 2.345(2) 2.354(5) 2.361(5) 2.358(5) 2.937(4) (x2) 3.099(2) (x2) 3.043(3) (x2) Na(1)-Se (mean) 303(8) Se(1)-Au-Se(1 ) Se(2)-Au-Se(2) Se( 1 )-Au-Se(2) Se( 1 )-Au-Se(2) Au-Se(1 )-Se(1 ) Au-Se(2)-Se(3) Se(2)-Se(3)-Se(4) Na(2)-Se(3) Na(2)-83(4) Na(2)-Se(4) Na(2)-Se(4) Na(2)-Se (mean) 130.0(0) 130.0(0) 90.9(1) (x2) 39.1(1) (x2) 1 10.7(1) 109.0(1) 101.4(2) 320(2) 294(1) 306(2) 295(1) 304(12) 125 6528 9:33 5? «52m m2 .0 9:055:28 5:26.600 .3 9:9“. $9.... Ema 9‘4 axz 3% Eu: 50m a 50m E3 E3 126 3.2.6. Structure of NaAuSea (VII) The structure of NaAuSe2 possess a covalently bonded [AuSezlnn' anionic layered network instead of the one-dimensional chain of [AuSe2]n"' even though its stoichiometry in formula units is same as that of KAuSe2. The structure of one layer of [AuSe21nn' is shown in Figure 30. The anionic [AuSezhn' layer represents a new structure type for an MX2 stoichiometry. The layers are made of polymerized [Aque2]Se4/2 dimeric units with a central rhombic [AuzsegP't core. The Au--Au distance at 3.730(2) A does not imply any significant Au--Au interactions in the layer. The geometry of the Au3+ center is slightly distorted square planar. Two Au-Se distances are observed; Au- Se(1) inside the central [Aque2]2+ core is short at 2.460(4) while Au-Se(2) outside the central core is slightly longer at 2.498(4) A. Selected bond distances and angles are given in Table 45. The [AuSe2]n"' layers form corrugated sheets that stack in phase in the crystallographic a-axis direction, as shown in Figure 31. The interlayer spacing is 6.745 A. There is one crystallographically distinct Na atom in the asymmetric unit. Charge balancing Na+ ions are distributed between layers with ionic interactions with Se atoms. The Na atom has distorted octahedral geometry. The shortest Na-Se distance is 292(1) A. 127 Figure 30. ORTEP representation of the layered structure of [AuSez]n"', looking perpendicular to the layer. 128 Table 45. Selected Bond Distances (A) and Angles (deg) in NaAuSe2 with Standard Deviations in Parentheses Au-Se(1 ) Au-Se(1 ) Au-Se(2) Au-Se(2) Au-Se (mean) Na-Se(1 ) Na—Se(2) 2.460(4) 2.459(3) 2503(4) 2439(3) 243(2) 304(7) (x3) 3.01(9) (x3) Na-Se (mean) 303(2) Se(1 )-Au-Se(1 ) Se(2)-Au-Se(2) Se(1 )-Au-Se(2) Se(1 )-Au-Se(2) Se(1)-Au-Se(2) Se(1)-Au-Se(2) Au-Se(1 )-Au Au-Se(2)-Au 81 .4(1) 93.1(1) 96.0(1) 170.3(1) 176.6(1) 89.6(1) 98.6(1) 102.9(1) 129 I; V \ . ‘\\§¢. w I l 3:" / Q ’ 3‘9"”? A " A -24” I” I" ‘f‘ A.( A\\ ‘\ \\\ \v ' ‘31., A ’ v / Figure 31. ORTEP representation of the unit cell of NaAuSez. 130 3.3. Structures The compounds isolated in this system show a rich structural chemistry involving the various polychalcogenide ligands. All compounds obtained in the ternary AlAu/Q (A=alkali metal) system can be grouped into two different categories based on the oxidation state of the central metal: (i) Au+ polychalcogenide compounds, and (ii) Au3+ (poly)chalcogenide compounds. The Au+ compounds, KAuSs, KAuSe5, and CsAuSea, are homologous to known zigzag chain compounds of KAuQ (Q=S, Se)8°. The anionic chains of these polychalcogenide compounds are viewed as derivatives of [AuQ]n"' chains80 in which 02' ligands have been replaced with polychalcogenides (852', Sesz',and 8e32' respectively). Alternatively, the structures of these compounds can be viewed as being derived from helical polymeric chains of sulfur (Sx) or selenium (Sex) by substitution of linear Au+ atoms for every sixth sulfur (or selenium) atom for (I) and (II) and every fourth atom for (Ill). However, the helical character of the Se chain is maintained only in the CsAuSea structure. In the KAuQs compounds short Au-Au contacts are observed at 2.963(1) A and 2.950(3) A for (l) and (II), respectively. This is a beautiful example of a well defined, isolated d10-d1o interactions68 in the solid state. In general, d10- d10 bonding contacts are not fully understood. Normally, the interaction of two- filled metal orbitals results in a nonbonding situation due to formation of a bonding and an equally antibonding state as shown in the following scheme. Of course, the result of such interaction is zero. 131 /$ wo\§ However, recent theoretical investigations into this issue have invoked a significant admixture of s and p character (from the empty 5 and p orbitals) into predominantly d metal orbitals.86 This hybridization gives rise simultaneously to a more bonding and a less antibonding M-M interaction and therefore an overall stabilization as shown in the following scheme. 132 \ A“ ‘\ antibonding 5d \ noninteracting d-d interaction with s and p ions alone mixing Two sets of intriguing Au3+ compounds, (i) IQAuSe13 and Na3AuSea, and (ii) KAuSez and NaAuSe2, exhibit cation size effect on structures and interesting redox chemistry associated with Au+IAu3+ ion in the alkali metal polyselenide fluxes. The anionic chains of [Au(Se3)(Se5)2]n3"' and [Au(Sez)(Se3)2]n3"' in (N) and (VI), respectively, can be viewed as the oxidative addition product of Sex2‘ (x=10, 6) ligands to the Au+ atoms of one-dimensional [AuSey]n"' (y=3,2) chains as shown in eq 3. [AuSesln"' + nSe1o2' ------- > [AU(Se3)(Ses)2ln3"' e93(a)- [AuSezlnn' + nSesz' ------- > [Au(Sez)(Sea)2]n3"' eq 3(b). 133 Yet, the hypothetical anionic chain of [AuSezjnn' in eq 3 (b) is not known. It is, however, another possible phase in the homologous series of compounds of [AuSexhn' (x=1, 3, 5 known) chain. (IV) and (VI) show very unusual binding modes such as monodentate ligation for the long polychalcogenide ligands. The usual binding mode of polychalcogenides (Sex2' x>2) known to date is at least bidentate, either bridging or chelating as shown in the previous chapter (see Table 2). Another example of these unusual monodentate ligation of 8e52- have been found in C34[Se16].87 In this compound a square planar 862+ atom is bonded by three Se52' chains, of which one act as bidentate chelate, while the other two act as monodentate ligands. In the structure of (IV) and (VI) alkali metal cations interact ionically with those monodentate polyselenide ligands to stabilize the unusual monodentate ligation. A second set of Au3+ compounds, (V) and (VII) with the same formula unit stoichiometry [AuSezln'l’ gives an excellent example of counterion size effect on structure (one-dimensional vs. two-dimensional). Substitution of Na+ for K+ in (V) would result in a significant decrease in cation partial volume and would bring the [AuSe]n"' chains close enough that considerable ooulombic interchain repulsions might develop, thus destabilizing the structure. These repulsions could be overcome and converted to attractive forces by combining the chains into layers through interchain Au-Se bonding. This structural relationship is shown in Figure 32. One can visualize Au-Se(2) bond breakage along the dotted line in the layer to form [Auzse2]SeZSe2/2 units and flexible [Aque2]Se2/2 units. 134 .E 5 .2693 6 22626 526 as 2, ea 5 .2695 .6 9262.6 6.3. i n .3893 .o 3.202% co Bozo cum 5:8 22:38 on. .0 5395852 02689.8 .3 2:9". 135 These [Auz86218e2/2 units can rotate to complete square planar coordination of the Au3+ ion by sharing Se atoms of [Auzse2]Se28e2/2 and produce one- dimensional chains of (VI). It will be interesting to determine any structural changes that occur in the [AuSe2]n"' unit with even smaller Li+ or larger Cs+ counter ions. These compounds have not been synthesized yet. However, one could predict a three-dimensional structure of a Li+ salt because the smaller Li+ would bring the layers in the Na+ salt closer together and rearrange the two-dimensional framework into a three-dimensional structure to overcome the ooulombic repulsions between the layers. Since one-dimensional chains are more flexible in accepting a larger cation, similar chain structure of KAuSe2 should remain unchanged for the 06+ salt. The investigation of ternary AIAu/Q systems yields an intriguing set of new compounds with novel structural architectures and underscores the significance of the polychalcogenide molten salt method for new materials synthesis. The slight but significant variation in the polyselenide melt composition and the isolation of both Au” and Au3+ polyselenide compounds highlight the complexity of these reaction mixtures and the delicate coordination and redox equilibria present among various Aum+ (m=1, 3) and Sexz' species. CHAPTER 4 Molten Salt Synthesis of Low-Dlmenslonal Ternary Chalcogenldes. Novel Structure Types In the All-Iglo System (A=K, Cs; 0=S, Se) 1 . Introduction We have demonstrated in previous chapters that by using alkali metal polychalcogenide fluxes as solvents and reagents at intermediate temperatures (e.g. 210-350 00), novel, low-dimensional solid compounds can be isolated in crystalline form. We believe that an enormous number of interesting and perhaps metastable compounds with low-dimensional structures occur at these temperatures (150~500 0C). Solid chalcogenides are of intense interest due to their useful electronic55 and catalytic27 properties. Therefore, new such materials are highly desirable. Ternary Ang/Q compounds (A: alkali metal) have been rarely explored. Prior to this work there was only one known structural class of compounds, AeHgQ4 (A=K, Rb; Q=S, Se)83. Therefore, we initiated the investigation of ternary A/Hg/Q systems using alkali metal polychalcogenide fluxes (AzQx) at intermediate temperatures. In this chapter, we illustrate that alkali metal polychalcogenide fluxes at 210~370 00 promote synthesis and crystal growth of new ternary compounds, AzHgaQ4 (A=Na, K, 136 137 Cs; Q=S, Se) and AzHng-I (A=K, Cs; Q=S, Se), which feature novel structure types. 2. Experlmental Sectlon 2.1 Reagents Chemicals in this work were used as obtained: mercury sulfide (HgS) powder, analytical reagent, .J. T. Baker Chemical Co., Phillipsburg, NJ; mercury selenide (HgSe) powder, -100 mesh, 99.9% purity, Cerac, Milwaukee, WI; sulfur powder, sublimed, .J. T. Baker Chemical Co., Phillipsburg, NJ; selenium powder, -100 mesh, 99.95% purity, Aldrich Chemical 00., Milwaukee, WI; potassium metal, analytical reagent, Mallinckrodt lnc., Paris, KY; cesium metal, 99.98% purity, AESAFI, Johnson Matthey, Seabrook, NH. 2.2. Physlcal Measurements Quantitative microprobe analysis was performed on a Jeol SSCF scanning electron microscopy equipped with Tracor Nothern TN5500 X-ray microanalysis attachment. Single crystals of each sample were carefully picked and mounted on an aluminum stub using conducting silver paint to help dissipate charges that develop on the sample surface during measurements. Energy Dispersive Spectra (EDS) were obtained using the following experimental set-up: X-ray detector position : 55 mm Working distance : 39 mm 138 Accelerating voltage : 20 KV Take-off angle : 27 deg Beam current : 200 picoamps Accumulation time : 100 seconds Window : Be A standardless quantitative analysis (SQ) program was used to analyze the X- ray spectra obtained. Since the selenium ratio is always underestimated due to an artifact of the program, a correction factor (x183), which was determined with a known HgSe compound, was used to evaluate the selenium percentage. The analysis reported is the average of four to six individual measurements on different crystals. 2.3. Synthesls Chemicals were measured and loaded in Pyrex tubes under a dry nitrogen atmosphere in a Vacuum Atmospheres Dri-Lab glovebox. Potassium monosulfide (K28) and alkali metal monoselenide (AZSe; A=K, Cs) were prepared in liquid ammonia from alkali metal and elemental sulfur (or selenium) in a 2:1 ratio. Dipotassium tetra(p2-suIfido)trimercurate(ll), K2Hgas4 (I) 0.165 g (1.5 mmol) of K28, 0.116 g (0.5 mmol) of H98, and 0.128 g (4.0 mmol) of S were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 220 0C for 99 hrs and cooled slowly to 50 0C at a rate of 2 oC/hr. Pale yellow transparent hexagonal crystals were obtained with small contamination of H98 by removing excess potassium polysulfides with 139 degassed dimethylformamide (DMF) under a N2 atmosphere (yield: 46 % based on the amount of Hg used). The product is not stable in water and decompose rapidly. The presence of K and Hg and S atoms in a large number of crystals was confirmed using the EDS/SEM system. Dlpotasslum tetramz-selenldonrlmercurate(II), K2H93504 (II) 0.118 g (1.75 mmol) of Kzse, 0.070 g (0.25 mmol) of HgSe, and 0.158 g (2.0 mmol) of Se were mixed together and loaded in a Pyrex tube which was flame- sealed under vacuum (~10'3 torr). The tube was placed in a computer- controlled furnace and heated at 250 0C for 99 hrs and cooled slowly to 50 0C at a rate of 2 oClhr. Red hexagonal crystals were obtained with small contamination of HgSe by removing excess potassium polyselenides with degassed DMF under a N2 atmosphere (yield: 53 % based on Hg used). The product is not stable in water and decompose rapidly. The presence of K and Hg and Se atoms in a large number of crystals was confirmed using the EDS/SEM system. cheslum tetra(p2-selenldo)trlmercurate(ll), Cszflgasu (III) 0.115 g (0.33 mmol) of Cszse, 0.047 g (0.17 mmol) of HgSe, and 0.105 g (1.33 mmol) of Se were mixed together and loaded in a Pyrex tube which was flame- sealed under vacuum (~10:3 torr). The tube was placed in a computer- controlled furnace and heated at 250 °C for 99 hrs and cooled slowly to 50 0C at a rate of 2 OClhr. Orange-yellow hexagonal crystals were obtained, with small contamination of HgSe, by removing excess cesium polyselenides with degassed DMF under a N2 atmosphere (yield: 58 °/o based on Hg used). The product is relatively stable in water in a short time of period, but decomposes in an hour. A quantitative microprobe analysis performed on a large number of 140 crystals with the EDS/SEM system gave an average composition of CS1.9H93.0363.6- Dlpotasslum hepta(p2-sulfldo)hexamercurate(ll), K2H9587 (IV) 0.055 g (0.5 mmol) of K2S, 0.203 g (0.87 mmol) of H98, and 0.064 g (2.0 mmol) of S were mixed together and loaded in a Pyrex tube which was flame- sealed under vacuum («40‘3 torr). The tube was placed in a computer- controlled furnace and heated at 370 °C for 99 hrs and cooled slowly to 50 °C at a rate of 2 °CIhr. Black needle-like crystals were obtained, with little contamination of red HgS crystals, by removing excess molten potassium polysulfides with water under a N2 atmosphere (yield: 72 °/o based on Hg used). The product was washed with ethanol and ether and vacuum dried. The product is not soluble in water and any common organic solvents. A quantitative microprobe analysis performed on a large number of crystals with the EDS/SEM system gave an average composition of K2,ngs,387,1. This compound can also be prepared by direct synthesis: 0.110 g (1.0 mmol) of K28 and 1.396 g (6 mmol) of HgS were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 375 0C for 7 days and cooled slowly to 50 °C at a rate of 2 oC/hr. Black powder of KzHgGS-I were obtained with little contamination of red HgS. It was identified with X-ray powder diffraction pattern. cheslum hepta(p2-selenldo)hexamercurate(ll), Cszl-IgGSe7 (V) This compound could be prepared only by direct synthesis: 0.086 g (0.25 mmol) of 05236 and 0.420 g (1.50 mmol) of HgSe were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The 141 tube was placed in a computer-controlled furnace and heated at 375 0C for 72 hrs and cooled slowly to 50 0C at a rate of 3 °CIhr. Black needle-shaped crystals were obtained. The product is not soluble in water and any common organic solvents. A quantitative microprobe analysis performed on a large number of crystals with the EDS/SEM system gave an average composition of 062.1H9603673- 2.4. x-ray Crystallographic Studies All compounds were examined by X-ray powder diffraction for the purpose of phase characterization and identification. The d-spacings for each compound were obtained from the powder pattern recorded on a Phillips XRG- 3000 computer-controlled powder diffractometer, operating at 40KV, 35 mA. Graphite monochromated Cu radiation was used. The d-spacings observed for the bulk materials of (l) and (II) matched well with those calculated from the single crystal X-ray structure analysis data. However, they showed extra peaks which correspond to small impurities of H98 and HgSe, respectively. The d- spacings observed for manually selected single crystals of (III), (IV), and (V) were compared, and found to be in accord, with the calculated d-spacings from the single crystal X-ray structure analysis data. The calculation of d-spacings were performed using the POWD10 program56. The results are summarized in Tables 46-50. 142 Table 46. Calculated and Observed X-ray Powder Diffraction Pattern of K2H93$4 H K L awed.) 60mm) lllmax (obs) 0 0 2 6.65 6.90 33.1 o 0 4 3.42 3.43 100 0 2 1 3.17 3.16 5.6 3 1 0 3.09 3.10 7.2 2 1 3 3.05 3.06 4.5 3 1 1 3.02 3.03 4.0 0 2 2 2.94 2.95 2.5 3 1 2 2.62 2.62 12.5 0 2 3 2.65 2.67 10.4 3 1 3 2.56 2.57 4.4 0 2 4 2.364 2.366 3.1 3 1 4 2.296 2.301 12.7 0 0 6 2.264 2.265 29.9 0 2 5 2.099 2.104 6.3 3 1 5 2.053 2.057 4.7 0 2 6 1.6720 1.6747 2.4 3 1 6 1.6366 1.6425 13.1 3 3 1 1.6354 1.6364 11.6 3 3 3 1.7164 1.7169 40.4 0 0 6 1.7132 1.7166 22.4 6 0 2 1.7046 0 4 1 1.6220 1.6265 3.2 0 4 2 1.5669 1.5646 1.7 6 0 4 1.5656 1.5694 1.6 143 Table 47. Calculated and Observed X-ray Powder Diffraction Pattern of K2H938e4 H K L 4216(7)) dam (A) l/Imax(obs.) 0 0 2 7.02 7.22 100 r 3 1 0 3.16 3.20 34.2 “ 0 2 2 3.05 3.09 14.6 3 1 2 2.90 2.90 30.4 0 2 3 2.74 2.77 26.1 ‘L 0 2 4 2.43 2.43 6.0 ’ 3 1 4 2.358 2.373 52.4 0 2 5 2.163 2.161 31.1 3 1 5 2.106 2.137 13.1 2 3 2 1.999 2.007 10.7 0 2 6 1.926 1.939 11.6 3 3 1 1.6961 1.914 26.6 3 1 6 1.6656 3 3 3 1.7729 1.760 34.6 0 0 6 1.7552 0 4 1 1.6635 1.667 5.0 0 4 2 1.6463 1.661 9.7 3 3 5(6 2 1) 1.5625 1.590 13.1 144 Table 48. Calculated and Observed X-ray Powder Diffraction Pattern of CszH93$e4 H K L deA) dowel.) l/lmax(obs.) 0 0 2 7.36 7.51 76.9 0 0 4 3.69 3.72 94.2 3 1 0 3.41 3.44 36.7 2 1 3 3.26 3.31 100 3 1 2 3.09 3.12 46.6 4 0 0 3.01 3.03 16.4 0 2 2 2.96 2.98 56.7 3 1 4 2.50 2.52 37.4 2 1 5 2.453 2.469 31.2 0 2 4 2.432 2.445 29.3 4 2 0 2.203 2.215 14.2 3 1 6 1.996 2.004 21.1 0 2 6 1.956 1.967 49.0 6 0 2 1.9374 1.9396 11.6 4 0 6 1.9060 1.9093 16.5 0 0 6 1.6463 1.6541 77.9 6 1 3 1.7667 1.7934 10.9 3 1 6 1.6237 1.6264 10.5 145 Table 49. Calculated and Observed X-ray Powder Diffraction Pattern of K2H96$7 H K L dam/i) dam/X) l/lmax(obs.) 2 0 0 6.90 6.99 7.5 2 1 0 6.17 6.24 42.6 3 1 0 4.36 4.40 16.6 4 0 0 3.45 3.47 45.1 2 1 1 3.40 3.43 54.6 4 1 0 3.34 3.37 100 3 3 0 3.25 3.27 43.0 2 2 1 3.13 3.14 66.2 4 2 0 3.06 3.10 13.1 3 0 1 3.05 3.06 14.4 3 2 1 2.79 2.80 21.9 4 3 0 2.76 2.77 67.9 5 1 0 2.70 2.72 16.2 5 2 0 2.56 2.57 36. 4 2 1 2.461 2.470 16.3 4 4 0 2.440 2.446 11.5 4 4 1 2.094 2.099 16.0 5 3 1 2.047 2.046 23.5 0 0 2 2.040 2.044 16.0 6 1 1 1.963 1.966 47.4 5 5 0 1.952 1.957 15.2 6 4 0 1.9144 1.9191 16.4 6 3 1 1.6374 1.6413 13.7 7 1 1 1.7610 1.7651 51.1 6 1 0 1.7123 1.7156 17.9 7 3 1 1.6565 1.6602 14.7 4 3 2 1.6407 1.6435 6.5 6 3 0 1.6157 1.6188 20.9 5 2 2 1.5962 1.5966 4.0 Table 50. Calculated and Observed X-ray Powder Diffraction Pattern of 146 082H96$e7 H K L dcaMA) dam/X) l/Imax(obs.) 2 1 0 6.46 6.56 9.5 4 0 0 3.62 3.66 21.4 2 1 1 3.56 3.61 16.0 4 1 0 3.51 3.54 42.9 3 3 0 3.41 3.44 43.9 2 2 1 3.29 3.32 100 3 1 1 3.14 3.16 11.6 3 2 1 2.94 2.96 29.2 4 3 0 2.90 2.92 27.3 5 2 0 2.69 2.71 10.4 4 2 1 2.59 2.60 11.1 4 4 0 2.56 2.56 14.8 5 3 0 2.467 2.50 6.4 5 0 1(4 3 1) 2.406 2.419 16.6 5 4 0 2.265 2.295 7.7 0 0 2 2.154 2.162 42.0 6 1 1 2.066 2.097 42.0 6 3 1 1.9325 1.9405 31.3 7 0 1 1.6673 1.6770 13.7 7 1 1(5 5 1) 1.6520 1.6600 26.6 (4 0 2) 6 4 1(3 3 2) 1.6226 1.6426 21.1 6 1 0 1.7991 1.6061 23.9 7 3 1 1.7419 1.7492 7.7 4 3 2 1.7294 1.7356 11.1 6 3 0 1.6976 1.7046 13.4 5 2 2 1.6622 1.6699 7.9 5 3 2 1.6263 1.6345 10.6 147 The X-ray single crystal data of K2H93S4, CszHgase4, K2H9667, and CszHgGSe-I were collected on a Nicolet P3 four-circle diffractometer with graphite monochromated Mo-KCl radiation using the 0-20 scan mode. The data for K2H93Se4 were collected on an Enraf-Nonius CAD4 diffractometer with Mo- Kcl radiation using the (1)-20 scan mode by Dr. M. Sabat at Northwestern University. Accurate unit cell parameters for all compounds were obtained from the least-squares refinement on the 20, m, x, and 1) values of several (20-25) machine-centered reflections. The stability of the experimental setup and crystal integrity were monitored by measuring three standard reflections periodically (every 100 reflections) during the data collection period. The data of (I) showed appreciable decay (34.3 % decay) due to decomposition of the crystal. The color of the crystal changed from yellow to light black during the data collection period. The data of (II), (III), (IV), and (V) did not show any appreciable decay. Two absorption corrections were applied to all data: an empirical absorption correction based on up scans for 3 reflections followed by a DIFABS57 correction. The structures were solved with direct methods using SHELXS-8658 and were refined with the SDP59 package of crystallographic programs on a VAXstation 2000 computer. The effect of secondary extinction was considered in the least-square refinement as an additional parameter due to the heavy atomic constituents of the crystals. All atoms except the sulfur atoms in K2H93S4 were refined anisotropically. Due to decomposition of the crystal during the data collection the sulfur atoms of KzHg3$4 could not be refined anisotropically. The complete data collection parameters and details of the structure solution and refinement for all compounds are given in Table 51. The final atomic coordinates, temperature factors and their estimated standard deviations are given in Tables 5256. 148 Table 51. Summary of Crystallographic Data for K2H9384, K2Hgase4, CSzHgaSe4, K2Hgss7, and C82H96887 compound I II III Formula K2Hg3S4 K2H93Se4 Cs2Hgase4 Formula weight 808.23 995.81 1 183.42 space group Pbcn Pbcn Pbcn a (A) 10.561 (5) 10.620(2) 12.047(4) b (A) 6.534(3) 6.763(1) 6.465(2) c(A) 13.706(2) 14.042(2) 14.771(6) 01 (deg) 90.0 90.0 90.0 8 (deg) 90.0 90.0 90.0 7 (deg) 90.0 90.0 90.0 Vol (A3), 2 945.8(7), 4 1030.6(5), 4 1150.4(7), 4 Temperature (°C) 23 -120 23 Crystal size (mm) 0.15x0.10x0.01 0.17x0.12x0.02 0.22x0.18x0.02 Radiation Mo-Ka Mo-Ka Mo-Ka u (Mo-Ka, cm'1) 502.3 593.2 586.1 Dcalc (g/cm3) 5.68 6.42 6.83 26mm, (deg) ' 50 50 46 Scan method 0/20 0/20 0/20 No. of data collected 2829 1088 1058 No. of unique data 831 903 900 No. of data used 326 567 319 (F02>30(F02)) No. of atoms 5 5 5 No. of variables 33 43 43 Phasing technique Direct methods Direct methods Direct methods Final RlFlw 5.7/6.3 773/836 5.5/62 Max. shift/esd 0.00 0.00 0.00 (last cycle) Extinction coefficient 1.58x10'8 2.06x10'8 3.15x10'8 149 Table 51. (cont'd) 00mm IV V Formula K2HgGS7 Cs2Hg5Se7 Formula weight 1506.19 2022.07 space group P-421m P42nm a (A) 13.605(6) 14.505(7) b(A) 13.605(6) 14. 505(7) c(A) 4.060(3) 4.306(2) 01 (deg) 90.0 90.0 8 (deg) 90.0 90.0 7 (deg) 90.0 90.0 Vol (A3), 2 776(1), 2 906(1), 2 Temperature (°C) 23 23 Crystal size (mm) 0.46x0.04x0.04 0.76x0.02x0.04 Radiation Mo-Ka Mo-Ka p (Mo-Kc, cm'1) 604.2 684.7 Dcalc (g/cm3) 6.43 7.41 20".” (deg) 50 44 Scan method 0/20 0/20 No. of data collected 1614 1440 No. of unique data 453 604 No. of data used 402 530 (F02>30(F02)) No. of atoms 6 6 No. of variables 42 41 Phasing technique Direct methods Direct methods Final R/Rw 3.1/3.6 3.6/4.0 Max. shift/esd 0.00 0.00 (last cycle) Extinction coefficient 1.50x10'7 1.11x10'7 150 Table 52. Fractional Atomic Coordinates and Beq Values for K2Hgas4 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Hg(1) 0.34148(9) 0.0852(3) 0.2582(1) 1 .48(3) Hg(2) 0 0.0637(5) 1/4 177(5) K 0.3788(6) 0.766(3) 0.501 8(8) 1 . 8(2) S(1) 0.3674(7) 0.175(2) 0.6416(6) 0.8(2)* 3(2) 0.3710(7) 0.362(2) 0.3678(8) 1 .1 (2)’ ‘3 B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[a2811 + b2822 + 02833 + ab(cos y)B12 + ac(cos (3)813 + bc(cos (1)823]. Starred atoms were refined isotropically. Table 53. Fractional Atomic Coordinates and Beq Values for KzHgase4 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Hg(1) 0.3383(2) 0.0889(3) 0.2586(2) 2. 74(4) Hg(2) 0 0.0674(6) 1/4 277(6) Se(1) 0.1338(5) 0.3239(9) 0.1382(4) 2. 2(1) 86(2) 0.3723(6) 0.3649(8) 0.3707(4) 2.3(1) K 0.118(1) 0.270(2) 0.501(1) 2.6(2) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (413)[a2811 + b25422 + 02833 + ab(cos 0812 + £1de 0813 + bc(cos a)8231. 151 Table 54. Fractional Atomic Coordinates and Beq Values for CszH93Se4 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Hg(1) 0.3309(2) 0.0051 (6) 0.2416(3) 3. 66(4) Hg(2) 0 -0.0123(9) 1/4 3.40(6) Cs 0.1203(3) 0.2656(4) 0.5105(3) 2.94(6) Se( 1) 0.1234(5) 0.2316(7) 0.1353(4) 2. 5(1) Se(2) 0.3604(4) 0.2918(7) 0.3541(5) 2.2(1) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[a2811 + b2822 + 62833 + ab(cos 10812 + actcos (3)813 + bc(cos 00823]. Table 55. Fractional Atomic Coordinates and Beq Values for KzHgss-I with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Hg(1) 0.56782(6) 0.341 1 5(7) 0.7608(4) 264(2) Hg(2) 0.90144(6) 0.599 0.7238(4) 214(2) K 0.6713(4) 0.829 0.764(3) 3. 5(1) S(1) 0.6965(4) 0.4650(4) 0.743(2) 1 .8(1 ) 3(2) 0 1/2 0.061 (4) 2. 8(2) 3(3) 0.81 39(4) 0.686 0.330(2) 2. 0(1 ) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[a2811 + 02822 + 02333 1' ab(cos Y)B12 1' 30(005 @313 1' (”(005 (1)323]- 152 Table 56. Fractional Atomic Coordinates and Baq Values for CszHgGSe-I with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Hg(1) 0.07503(9) 0.3466(1) 0.056 271(2) Hg(2) 0.1 0077(9) 0.101 0.4687(5) 242(2) Cs -0.1666(2) 0.169 0.063(1) 266(4) 36(1) 0.0200(2) 0.6935(2) 0.556(1 ) 1 . 68(6) 86(2) "2 “2 0.285(2) 2. 28(9) 36(3) 0.3168(2) 0.683 0. 548(1 ) 1 .66(5) ‘3 B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[a2811 + b2822 + 02833 + ab(cos 10812 + £2de (3)813 + boloos 008231- 153 3. Results and Discusslon 3.1. Synthesls The synthesis of K2H9384, K2H938e4, CszH93Se4, and K2H9687 has been readily accomplished at the intermediate temperature range (220~370 0C) using alkali metal polychalcogenide fluxes as solvent and reagents as shown in eq 1. nAzO + mHgQ + qO --------- > AaHgbOy + Asz eqt. ln eq 1 A represents an alkali metal (K or Cs) and O a chalcogen atom (S or Se). Each mixture formed a complete melt at the temperature employed. After reaction, excess alkali metal polychalcogenide flux was easily removed from the product by washing with water or DMF. (l) and (II) are not stable in water and decompose rapidly in moist air and light to form black HgQ. Although (III), (IV), and (V) are relatively stable in the air and moisture for several days, (Ill) decomposes slowly to form black HgQ. Preparation of Cszngsey was unsuccessful in polychalcogenide flux (CszSex). However, direct synthesis at 375 0C using stoichiometric amounts of CszSe and HgSe in a 1:6 ratio yielded black crystals of CszH968e7 as shown in eq 2. CsQSe + 6 HgSe -------- > CS2H95S97 eq 2. An attempt to synthesize K2H95$7 by direct combinations of K28 and HgS in a 1:6 ratio at 375 0C yielded only microcrystalline K2H95S7. 154 In our initial investigations of the AIHng (A=alkali metal; O=S, Se) systems, we used Hg metal as a reactant. However, elemental mercury did not react completely. A surface-only reaction cover the Hg bead with ”crust" of product which prevented further reaction. When we substituted finely powdered HgO for mercury metal we obtained more homogeneous mixing and complete reaction. The compounds of AzHgaQ4 (A=K, Cs; Q=S, Se) were isolated in every reactant ratio from 211/8 to 5/1/8 (AzQ/HgQ/Q). Among them, the 3/1/8 ratio for (l) and (II), and the 2/1/8 ratio for (lll) gave better yield with larger single crystals suitable for a X-ray single crystal study. However, contamination of the HgO phase seen in every reactant ratio could not be avoided even by changing the reaction temperatures from 220 °C to 350 0C. Use of higher temperatures gave more HgO. The amount of H90 impurity is estimated roughly to be 10 % in (l), 20 % in (II), and 10 % in (lll). During the preparation of K2Hg3S4 we found a few black needle crystals of a new phase in the higher ratio of K28 (4~511l8). This phase was found to be Kgngs-z and was also observed in the preliminary reaction with mercury metal. Subsequently, K2H9687 was prepared at 370 °C in better yield and as large single crystals, up to 5 mm, by increasing the H98 ratio to 2135/8. Even though K2H93S4 formation could be completely suppressed using this reactant ratio, HgS was always present in small amounts. We were not successful in preparing the Se-analogue of K2HgGS7 either in Kzsex flux or by direct reaction of K2Se with HgSe. This is probably because of the smaller size of the K+ ions compared to enlarged size of Hg-Se tunnels created by the Se substitution. The size of the tunnel is too big for the smaller K+ ions and requires larger cations such as Cs+ ions. In fact, the Cs salt of Se-analogue, CSQHQGSe7 was prepared from the direct synthetic method. 155 During this study we wondered why no polychalcogenide compounds were obtained. We do not have a clear answer to this, but it may be explained in terms of the comparable charges on Hg2+ ions and monochalcogenide 02' ions which would form more energetically favorable lattice. However, one possible approach to obtain polychalcogenide compounds could be to provide polychalcogenide fluxes with much longer chains in combination with lower reaction temperature. This could increase the concentration of polychalcogenide species vs monochalcogenide species in the flux and possibly force the reaction equilibrium towards formation of polychalcogenide compounds. 3.2. Description of Structures 3.2.1. Structure of K2Hgas4(l), K2H93$e4(ll), and ngflgase4(lll) (I), (II), and (III) are isostructural. They possess centrosymmetric one- dimensional [Hg304]n2"' chains. Two views of the unit cell of A2Hg3O4 are shown in Figure 33. The [HgaQ4ln2'l' chains run parallel to the crystallographic b-axis. The make-up of this chain can be regarded as a one-dimensional assembly of distorted tetrahedral [HgQ4]6' building blocks connected by two- coordinate H92+ ions as shown in Figure 34. Alternatively, it can be viewed as a one-dimensional spiro-polymer of eight-membered Hg4O4 rings. There are two crystallographically distinct Hg atoms in the asymmetric unit. The Hg(1) atom has linear geometry, while Hg(2) atom, situated on a crystallographic 2- fold axis, has distorted tetrahedral geometry. The angles of the linear Hg(1) atoms at 165.3(3), and 164.2(2), and 158.8(2) deg for (I), (II), and (Ill), 156 Hg(1) 0(2) 0(1) H9(2) (B) (A) .8, Se): Figure 33. Two axial views of the unit cell of AzHgaQ4 (A=K, Cs; (A) b-axial view and (B) projection view along (001) plane. 157 0.55.6. 2:86. 5.: 62.403:— 5 920:5 5565.25-05 05 5 m5; 93 .3 2:9“. Ea . f... . f... «a»: AfiNUvU—u— ,v 9v 7' a Ea a CC 158 respectively, are smaller than that found in the linear Hg atoms (172 deg) in HgS39. These are due to the short Hg(1)-0(2) contacts between the chains along (001) plane at 3.06 (1) A, 3.159(6) A, and 3.156(6) A for (I), (II), and (Ill), respectively. There are two sets of long and short Hg-Q bonds in this structure type associated with tetrahedral and quasi-linear H92+ centers. The average Hg-O bond distances are (a) for quasi-linear Hg(1) atom: 237(1) A, 2.48(1) A, and 247(6) A in (I), (II), and (Ill), respectively; (b) for tetrahedral Hg(2) atom: 257(1)A, 2.66(1)A, and 2.68(8)A in (I), (II), and (Ill), respectively . Selected bond distances and angles are given in tables 57-59 for (I), (II), and (Ill) respectively. There is one crystallographically distinct alkali metal in the asymmetric unit of each compound. The charge balancing alkali metal cations, A+ ions (A=K, Cs) are distributed between the chains. They participate in ionic interactions with the chalcogenides. The K+ ions are surrounded by six chalcogen atoms in the range of 3.22(2)-3.35(1) A and 3.30(2)-3.37(1) A in (l) and (II) respectively, while Cs+ ions in (III) are surrounded by seven selenium atoms in the range of 3.596(7)-3.844(7) A. 159 Table 57. Selected Bond Distances (A) and Angles (deg) in K2Hgas4 with Standard Deviations in Parentheses Hg(1)-S(1) Hg(1)-8(2) Hg(1)-S (mean) Hg(2)-S(1) Hg(2)-8(2) Hg(1)-S (mean) Hg(1)-S(1) Hg(1)-S(1) K-S(1) K-S(1) K—S(1) K-S (mean) 236(1) 236(1) 237(1) 258(1) (x2) 2.56(1) (x2) 2. 57(1) 3.08 (1) 3.15 (2) 329(2) 335(1) 326(1) 329(4) S(ll-H9(1)-S(2) S(ll-H9(2)-S(1) S(1)-H9(2)-S(2) S(1l-H9(2)-S(2) 8(2)-Hg(2)-S(2) H90 )-S(1)-H9(2) Hg(1)-S(2l-H9(2) K-S(2) K-S(2) K-S(2) 165.3(3) 104.7(3) 105.7(3) (x2) 114.9(3) (x2) 1 1 1 .0(4) 96.5(4) 95.8(4) 3.22(2) 330(1) 327(1) 160 Table 58. Selected Bond Distances (A) and Angles (deg) in K2H93Se4 with Standard Deviations in Parentheses Hg(1)-Se(1) 2.466(6) Se(1)-Hg(1)-Se(2) 164.2(2) Hg(1)-Se(2) 2.473(6) Se(1)-Hg(2)-Se(2) 115.7(2) (x2) Hg(1)-Se (mean) 2.480(9) Se(1)-Hg(2)-Se(2) 104.4(2) (x2) Se(2)-Hg(2)-Se(2) 1 10.7(2) Hg(2)-Se(1) 2.671(6) (x2) Se(1)-Hg(2)-Se(1) 106.2(2) Hg(2)-Se(2) 2.657(6) (x2) Hg(1 )-Se(1)-Hg(2) 95.7(2) Hg(2)-Se (mean) 2.664(8) Hg(1)-Se(2)-Hg(2) 95.8(2) Hg(1)-Se(1) 3.159(6) Hg(1)-Se(1) 3.210(6) K-Se(1) 337(1) (x3) K-Se(2) 3.36(1 ) K-Se(2) 330(2) K-Se(2) 3.35(1 ) K-Se (mean) 3.35(3) 161 Table 59. Selected Bond Distances (A) and Angles (deg) in CszHgase4 with Standard Deviations in Parentheses Hg(1)-Se(1) Hg(1)-Se(2) 2.429(7) 2.515(7) Hg(1)-Se (mean) 247(6) Hg(2)-Se(1 ) Hg(2)-Se(2) 2.751(6) (x2) 2.608(6) (x2) Hg(2)-Se (mean) 268(8) Hg(1)-Se(1) Hg(1)-Se(2) Cs-Se( 1 ) Cs-Se(1 ) Cs-Se(1 ) Cs-Se(1 ) 3.158 (6) 3.295 (6) 3.596 (7) 3.647 (7) 3.706 (6) 3.737 (6) Se(1)-Hg(1)-Se(2) Se(1)-Hg(2)-Se(1) Se(1)-Hg(2)-Se(2) Se(1)-Hg(2)-Se(2) Se(2)-Hg(2)-Se(2) Hg(1)-Se(1l-H9(2) Hg(1 )-Se(2)-H9(2) Cs-Se(2) Cs-Se(2) Cs-Se(2) Cs-Se (mean) 1 56.6(2) 1 10.1 (2) 105.3(2) (x2) 107.0(2) (x2) 121 9(3) 962(2) 93.5(2) 3.706 (5) 3.734 (7) 3.644 (7) 3.66(16) 162 3.2.2. Structures of K2HgoS-1 (IV) and 092H95867 (V) The structures of K2H96S7 and CS2HgGSe7 are unique and possess three-dimensional frameworks with one-dimensional tunnels. The stereoviews of the unit cells of (IV) and (V) are shown in Figures 35 and 36, respectively. They are composed of two crystallographically distinct linear and tetrahedral ng+ ions bridged by three crystallographically distinct 02' ions. Even though they crystallized with different space groups (P-421 m for (IV), P42nm for (V)), the basic make-up of the three-dimensional framework of both compounds is the same. Every asymmetric atom has same point symmetry in both space group. Figure 37 shows the unit cells of (IV) and (V). The major difference between the two structures is the existing symmetry element, four-fold inversion axis (-4) in (IV) vs for-fold screw axis (42) in (V), at the center of the 8-membered empty channel 163 Figure 35. Stereoview of the unit cell of K2HgSS7. 164 Flgure 36. Stereoview of the unit cell of C62HgSSe7. 20050.50 50.0 5. 8 .0 NE 9 0055 500 00: 6v .0 :00 «E: 05 .0 :65 05. 608100 6 =8 =5 05 a: 6.6 58:2 6 =8 2.5 21$ 6 6.668060. 3:00 .8 2.6m .6 i O a , LIV.» § ..-/«‘2 £0 \ . .2\ . s 07 v... . ... O 0 3 J tv ctzli/Ww. 000 //\. _l.\\. 80 ... ...I/ . . . 166 There are two sets of easily recognizable one-dimensional tunnels running parallel to the crystallographic c-axis. A set of empty, narrow tunnels with an octagonal cross section is composed of distorted tetrahedral Hg(1) and trigonal pyramidal 0(1) ions. The diameters of these tunnels are ca. 4.769(1) A and 4.696(2) A, corresponding to Hg(1)--Hg(1) distance, for (IV) and (V) respectively. A second set of ”stuffed", wider tunnels has a 12-membered ring cross section in which both tetrahedral Hg(2) and linear Hg(1) atoms are held together by triply and doubly bridging 02’ ions. The geometry of doubly bridging 0(2) and triply bridging 0(1) atoms are V-shaped and trigonal pyramidal, respectively. However, the other triply bridging Q(3) ion has an unusual T-type coordination as shown below: Hg(1) H9(1) \Q(3)/ H9(2) A mirror plane passes through the Hg(2)-0(3) bond, bisecting the Hg(1)-0(3)- Hg(1) angle. In (IV) the S(3) atom is on the trigonal plane of the three Hg atoms, but in (V) Se(3) atoms are slightly displaced from the plane. This displacement imposes lower symmetry on the framework of (V). The Hg(1)-S(3)-Hg(1) and Hg(1)-S(3)-Hg(2) angles in (IV) are 158.1(3)° and 100.9(2)o respectively. The Hg(1)-Se(3)-Hg(1) and Hg(1)-Se(3)-Hg(2) angles in (V) are 156.2(1)° and 97.5(1)o respectively. The Hg(1)-0(3) bonds are unusually long at 2.718(4) A, and 2.679(2) A in (IV) and (V), respectively, while the Hg(2)-0(3) bonds are 167 normal (for two-coordinated l-ng+ ion) at 2.345(6) A. and 2.477(5) A for (IV) and (V), respectively. As a result, there are two kinds of Hg-O bond distances associated with tetrahedral or linear coordination. The average Hg-O bond distances of tetrahedral Hg(1) centers are long at 2.57 (10) A and 2.67 (13) A and those of linear Hg(2) centers are short at 2.36 (1) A and 2.477(1) A in (IV) and (V) respectively. These are similar to those observed in (I), (II), and (Ill). The angles of linear Hg(2) atom in (IV) and (V) are 172.4(4)°, and 166.4(2)° respectively. Selected bond distances and angles are given in Tables 60 and 61 for (IV) and (V), respectively. There is one crystallographically distinct alkali metal in the asymmetric units of (IV) and (V). The alkali ions (K+, Cs+) are found inserted in the center of the large 12-membered tunnels, interacting with the chalcogenide lone pairs of electrons that are directed toward the tunnel center. The alkali ions are surrounded by seven chalcogen atoms in the range of 3.30(1)-3.62(1) A for K+ in (IV) and 3.626(4)-3.764(5) A for Cs+ in (V). 168 Table 60. Selected Bond Distances (A) and Angles (deg) in K2H9587 with Standard Deviations in Parentheses Hg(1)-S(1) 2.467(5) S(1)-Hg(1)-S(1) 121 .2(2) Hg(1)-S(1) 2.552(8) S(1)-Hg(1)-S(1) 124.8(3) Hg(1)-S(1) 2.527(8) S(1)-Hg(1)-S(1) 106.9(2) Hg(1)-S(3) 2.718(4) S(1)-Hg(1)-S(3) 96.3(2) Hg(1)-S (mean) 2.57(11) S(1)-Hg(1)-S(3) 93.7(2) S(1)-Hg(1)-S(3) 106.5(2) Hg(2)-8(2) 2.366(9) S(2)-Hg(2)-S(3) 172.4(4) Hg(2)-S(3) 2.345(8) Hg(1)-S(1)-Hg(1) 105.2(2) Hg(2)-S (mean) 2.36(15) Hg(1)-S(1)-Hg(1) 106.9(2) Hg(1)-S(1)-Hg(1) 102.6(3) Hg(2)-S(2)-Hg(2) 108.8(6) Hg(1)-S(3)-Hg(1) 158.1(3) Hg(1)-S(3)-Hg(2) 100.9(2) (x2) K-S(1) 334(1) (x2) K-S(3) 330(1) K-S(1) 3.30(1) (x2) K-S(3) 362(1) K-S(2) 3.420(6) K-S (mean) 339(12) 169 Table 61. Selected Bond Distances (A) and Angles (deg) in CS2HgGSe7 with Standard Deviations in Parentheses Hg(1)-Se(1) 2.637(4) Se(1)-Hg(1)-Se(1) 121.3(1) Hg(1)-Se(1) 2.622(4) Se(1)-Hg(1)-Se(1) 121.9(1) Hg(1)-Se(1) 2.564(3) Se(1)-Hg(1)-Se(1) 1 10.0(1) Hg(1)-Se(3) 2.870(2) Se(1 )-Hg(1)-Se(3) 94.5(1 ) Hg(1)-Se (mean) 2.67(13) Se(1)-Hg(1)-Se(3) 95.9(1) Se(1)-Hg(1)-Se(3) 104.8(1) Hg(2)-Se(2) 2.476(4) Se(2)-Hg(2)-Se(3) 166.4(2) Hg(2)-Se(3) 2.477(5) Hg(1)-Se(1)-Hg(1) 1 10.0(1) Hg(2)-Se (mean) 2.477(1) Hg(1)-Se(1)-Hg(1) 103.7(1) Hg(1)-Se(1)-Hg(1) 103.2(1) Hg(2)-Se(2)-Hg(2) 1 13.2(3) Hg(1)-Se(3)-Hg(1) 156.2(1) Hg(1)-Se(3)-Hg(2) 97.5(1) (x2) Cs-Se(1) 3.664 (4) (x2) Cs-Se(3) 3.764 (5) Cs-Se(1) 3.626 (4) (x2) Cs-Se(3) 3.690 (5) Cs-Se(2) 3.660(3) Cs-Se (mean) 367(5) 170 3.3. Structure The compounds prepared in ternary AIHg/Q systems ( A: K, Cs; Q=S. Se) , A2HgaQ4 and A2Hng7 can be regarded as members of a new general family with the chemical formula (A2Q)n(HgQ)m (n=1, m=3 in (I), (II), and (Ill); n=1, m=6 in (IV) and (V)). These structures can be viewed as deriving from the successive dismantling of the three-dimensional zinc-blende structure of H90 (Q=S, Se) by the various amounts of added A20 . The three-dimensional zinc- blende structure (see Figure 1) of H90 contains tetrahedral H92+ centers and tetrahedral 114-02' atoms with Hg-Q bond distances at 2.53 A and 2.63 A in H9889 and HgSe90, respectively. This three-dimensional framework of H90 is broken up as it tries to accommodate added A20. The 114-coordination of 02' atoms is reduced to pg- and 112-coordination in [Hg607]n2"' and to only (12- coordination in [HgaQ4Jn2"'. Another phase that also belongs to this homologous family is A6HgO433, (A2Q)3(HgQ)1, which features discrete tetrahedral [HgQ4]6' units. This compound can be considered as a complete breakup of the three dimensional framework of H90 into individual [HgO4]5' units. This is very similar to the successive breakup of the structures of the main group elements (e.g., Si, P) that ensues upon reduction with electro-positive metals to form the familiar Zintl phases91. Recently some more examples of successive dismantling of the three-dimensional zinc-blende framework of CdS has been discovered in our group92. The compounds K2Cd283 and K2Cd384, which feature tetrahedral [CdS4]5' units and p3- and 112-coordination of 32', make the whole new family of (K2S)n(CdS)m (n=1, m=2 and n=1, m=3 respectively). They are reminiscent of what we have seen in the All-lg/O system. However, isostructural compounds in both systems are not yet available even though the stoichiometry of the formula unit of K2Cd3$4 is the same as that of 171 A2Hgso4. This is probably because the Hg2+ atoms can afford two favorable coordination environments (linear and tetrahedral) while Cd2+ can afford only one favorable tetrahedral environment. In fact, the compound K2Cd3$4 shows the structural similarity to A2HgaQ4. In K2Cd384, the tetrahedral Cd2+ atoms occupying the linear H92+ site of the [H9304]n2"' chain extend its linear coordination into tetrahedral coordination through an interchain Cd-S bond, producing a two-dimensional layer. Based on the +2 formal charges on Hg atoms, and assuming -2 formal charges on sulfide atoms, all compounds are expected to have a completely filled valence band due to the d10 electronic configuration of ng+and should exhibit semiconducting behavior. Among them, three-dimensional compounds K2Hg5$7 and CS2HgSSe7 are expected to have smaller band gaps than the one-dimensional compounds of (I), (II), and (Ill) because its three-dimensional characteristics increases overlap of the band orbitals. We tried to measure the band gaps of (IV) and (V) using FT-IR spectroscopy in the mid- and near-IR region. However, in preliminary measurements we did not observe any spectral absorptions in those regions. Work to study their charge transport properties is in progress. This study in the ngO systems showed that the three-dimensional structure of H90 is tractable and can accommodate various amounts of alkali metal monochalcogenides. It is likely that the A2O/HgO system constitutes an infinitely adaptive pair similar to the (ZnS)n(ln2S3)m93 and (BaS)n(FeS)m94 systems. Work to identify other members of this family is in progress. CHAPTER 5 Synthesls and Characterizatlon of Layered Compounds of K2Cu5Tes and NaCuTe 1. Introductlon Since the discovery of high temperature superconductivity, mixed- valence Cu compounds have received considerable attention in the hope of discovering related materials with even higher transition temperatures. Although mixed-valence solid Cu-chalcogenide compounds have been known for some time, they remain relatively rare. For example, the only mixed-valence compounds known so far are NaaCU4S463, ACU403 (A=K, Rb, Cs, Tl; O=S, Se)“, A3CU3Ss (A=K, Fib)71 , A3CUBSes (A=Rb, CS)95, 03200586496. TlCuzOz (Q=S, Se)97, TlCu6S498, and Tl5Cu14Seto99. Even though the a- prioridesign of a high-Tc superconducting material is not yet possible, it is still worthwhile to pursue the synthesis of mixed valence Cu compounds since such electronic features generally play an important role for high electrical conductivity. Furthermore, compared to ternary sulfides and selenides, tellurides are even more rare.6.7.79(d).100 To the best of our knowledge, there are only four known phases NaCusTe2101, KCuaTe2102, and ACuTe(A=Na, K)74 in the ternary A/CulT e (A=alkali metal) systems, none of which have a mixed Cu oxidation state. With this rationale in mind, we decided to use the now 172 173 proven polychalcogenide flux method50'54 to explore new such compounds. In our hands, this synthetic method has been quite successful in preparing novel structural types of (poly)sulfides and (poly)selenides of various transition- metals.5°-53t54 (see previous chapters) Thus, we extended it to the telluride systems where we were able to synthesize the layered materials, chusTes and NaCuTe. The first is a novel mixed-valence Cu/l' e compound which possesses high metallic conductivity. The latter has been synthesized earlier at 800 0C by direct combination of the elements.74 Here we demonstrate a more convenient lower temperature synthetic method and its accurate crystal structure. 2. Experimental Section 2.1 Reagents Chemicals in this work were used as obtained: copper powder, -325 mesh, 99.95% purity, Cerac, Milwaukee, WI; tellurium powder, -100 mesh, 99.95% purity, Aldrich Chemical 00., Milwaukee, WI; potassium and sodium metal, analytical reagent, Mallinckrodt lnc., Paris, KY. 2.2. Physical Measurements Magnetic susceptibility measurements in the temperature range from 2 K to 300 K at 5 k6 were performed on a MPMS Quantum Design SQUID magnetometer. Single crystals of K2CU5Te5 were manually selected for measurements. They were used without grinding as random oriented single crystals. The data were corrected for diamagnetic contributions of the sample 174 holder. To obtain molar susceptibility the corrections for ion-core diamagnetic contributions from atomic constituents were made using the values tabulated by Mulay103. The magnetization of K2CU5T65 was examined at 5 K as a function of applied field from 250 G to 7 kG and was found to vary linearly with the applied field. Four-probe dc resistivity and thermoelectric power data for K2CU5Te5 were provided by Prof. Carl C. Kannewurf (Northwestern University) over the temperature range 5 K to 300 K. A computer automated measurement system was employed to obtain thermopower and resistivity data with both the current and thermal gradient applied along the needle axis. For all measurements electrode connections to the small single crystals were made with the use of 25 and 60 pm gold wires (and gold bonding paste. Quantitative microprobe analysis of the compounds were performed on a Jeol SSCF scanning electron microscope equipped with Tracor Nothern TN 5500 X-ray microanalysis attachment. Single crystals of each sample were carefully picked and mounted on an aluminum stub using conducting silver paint to help dissipate charges that developed on the sample surface during measurements. Energy Dispersive Spectra (EDS) were obtained using the following experimental set-up: X-ray detector position : 55 mm Working distance : 39 mm Accelerating voltage : 20 KV Take-off angle : 27 deg Beam current : 200 picoamps Accumulation time : 100 seconds Window : Be 175 A standardless quantitative analysis (SQ) program was used to analyze the X- ray spectra obtained. Since the Cu ratio is always overestimated due to the contribution of system Cu peaks, a correction factor (x 0.73), which was determined by calibrating with known KlCu/I’e compounds, was used to evaluate the Cu percentage. 2.3. Synthesis Chemicals were measured and loaded in Pyrex tubes under a dry nitrogen atmosphere in a Vacuum Atmospheres Dri-Lab glovebox. Potassium monotelluride (K2Te) and sodium monotelluride (Na2Te) were prepared in liquid ammonia from alkali metal and elemental tellurium in a 2:1 ratio. Dipotassium (pa-dltelluro)trls(p4-telluro)pentacuprate(l,II), K2CusTes (I) 0.309 g (1.5 mmol) of KzTe, 0.032 g (0.5 mmol) of Cu and 0.508 g (4.0 mmol) of Te were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 350 °C for 3 days and cooled slowly to 50 0C at a rate of 2 oC/hr. Black parallelepiped crystals were obtained by removing excess molten potassium polytellurides (K2Tex) with dimethylformamide (DMF) under a N2 atmosphere (yield: 77 % based on the Cu metal used). A quantitative microprobe analysis performed on a large number of crystals with the EDS/SEM system gave an average composition of KCU24Te26. Sodium 0:4-telluro)cuprate(l), NaCuTe(ll) 0.434 g (2.5 mmol) of Na2Te, 0.032 g (0.5 mmol) of Cu and 0.508 g (4.0 mmol) of Te were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum 176 (N103 torr). The tube was placed in a computer-controlled furnace and heated at 400 °C for 5 days and slowly cooled to 260 °C at a rate of 2 °CIhr, then to 140 0C at a rate 3 01hr, and to 50 0C at a rate of 10 °CIhr. Black plate-like crystals were obtained by removing excess molten potassium polytellurides (Na2Tex) with DMF under a N2 atmosphere (yield: 43 % based on the Cu metal used). A quantitative microprobe analysis performed on a large number of crystals with the EDS/SEM system gave an average composition of N31.OCU1.0T91.1- 2.4. X-ray Crystallographic Studies Each compound was examined by X-ray powder diffraction for the purpose of phase characterization and identification. The d-spacings for each compound were obtained from the powder pattern recorded on a Phillips XRG- 3000 computer-controlled powder diffractometer, operating at 40KV, 35 mA. Graphite monochromated Cu radiation was used. To verify product homogeneity, the d-spacings observed for the bulk materials were compared, and found to be in accord, with those calculated from the single crystal X-ray structure analysis data. The calculation of d-spacings was performed using the POWD10 program56. The results are summarized in Tables 62 and 63. 177 Table 62. Calculated and Observed X-ray Powder Diffraction Pattern of K20U5Tes H K L dm(A) dobs (A) l/lmax(obs.) 0 2 0 6.00 6.06 44.9 0 2 1 7.35 7.41 7.5 0 0 4 4.63 4.67 6.1 1 1 0 4.00 4.02 22.5 1 1 1 3.90 3.93 15.3 0 4 2 3.67 3.71 6.1 0 2 5 3.36 3.36 56.0 1 3 0 3.27 3.27 44.2 1 3 1 3.22 3.23 31.6 0 0 6 3.09 3.10 17.6 0 4 4(1 1 4) 3.03 3.04 46.4 0 2 6 2.88 2.69 24.2 1 3 4 2.67 2.66 16.2 1 3 5 2.45 2.450 32.5 1 5 2 2.440 1 5 3 2.341 2.350 15.3 1 3 6 2.244 2.251 34.4 1 5 4 2.220 2.226 44.2 2 0 0 2.065 2.069 26.9 2 2 0(0 2 9) 2.000 2.000 33.7 1 7 2 . 1.955 1.962 100 1 5 7 1.6291 1.6346 26.9 2 2 5 1.7597 1.7635 14.4 1 5 6 1.7065 1.7095 16.1 2 4 4 1.7060 1.6967 6.6 1 7 7 1.5959 1.6007 10.5 1 9 3 1.5790 1.5600 9.4 178 Table 63. Calculated and Observed X-ray Powder Diffraction Pattern of NaCuTe H K L dcak, (A) do),$ (A) l/lmax(obs.) 0 0 1 7.13 7.03 54.7 1 0 1 3.73 3.73 100 0 0 2 3.56 3.56 41.4 1 0 2 2.76 2.76 22.9 1 1 2 2.341 2.333 72.1 200 2m5 1 0 3 2.090 2.066 65.3 1 1 3 1.6674 1.8887 45.0 0 0 4 1.7627 1.7610 79.2 1 0 4 1.6518 1.6502 39.7 179 The X-ray single crystal data of K2CU5Te5 and NaCuTe were collected on a Rigaku AFCGS diffractometer with graphite monochromated Mo-Ka radiation using the (0'28 scan mode. Accurate unit cell parameters for all compounds were obtained from the least-squares refinement of the 20, to, x, and 0 values of 20-25 machine-centered reflections. The stability of the experimental setup and crystal integrity were monitored by measuring three standard reflections periodically (every 100 reflections) during data collection. The intensities did not show any appreciable decay. Two absorption corrections were applied to the data of K2CU5T65 and NaCuTe: an empirical absorption correction based on 11: scans for 3 reflections followed by a DIFABS57 correction. The structures of both compounds were solved with direct methods using SHELXS-8658 and were refined with the TEXSAN60 package of crystallographic programs. All calculations were performed on a VAXstation 3100 computer. All atoms were refined anisotropically. The complete data collection parameters and details of the structure solution and refinement for (I) and (II) are given in Table 64. The final atomic coordinates, temperature factors and their estimated standard deviations are given in Tables 65 and 66. 180 Table 64. Summary of Crystallographic Data for K2Cu5Te5 and NaCuTe commund I II Formula K2Cu5Te5 NaCuTe Formula weight 1033.93 214.14 space group Cmcm P4/nmm (2nd. setting) a (A) 4.130(3) 4.3913(9) c(A) 16.004(2) 4.3913(9) c (A) 16.533(3) 7.131(2) (1 (deg) 90.0 90.0 8 (deg) 90.0 90.0 7 (deg) 90.0 90.0 Vol (A3), 2 1225.0(9), 4 137.5(1), 2 Temperature (°C) 23 23 Crystal size (mm) 0.70x0.13x0.10 0.39x0.13x0.03 Radiation Mo-Ka Mo-Kot p. (Mo-Kat, cm'1) 209.2 181.9 Dcalc (glcm3) 5.61 5.17 20",” (deg) 50 60 Scan method (1)/20 0120 No. of data collected 2238 495 No. of unique data 663 154 No. of data used 569 134 (F02>30(F02)) No. of atoms 7 3 No. of variables 40 9 Phasing technique Direct methods Direct methods Final RIRw 2513.4 2.8I3.4 Max. shift/esd 0.00 0.00 (last cycle) Extinction coefficient 6.43x10'7 NIA 181 Table 65. Fractional Atomic Coordinates and Beq Values for K2CU5T65 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 16(1) 0 0.1281 1(3) 0.51974(4) 1.30(3) T6(2) 1/2 0.30524(3) 0.67015(4) 1 .44(3) Te(3) 0 0.07736(5) 3/4 132(4) CU(1) 0 0.19941 (7) 0.64842(8) 1 .96(6) CU(2) 0 0.27787(7) 0.46035(8) 2. 26(6) CU(3) 1/2 0.1698(1) 3/4 1 .80(7) K(1) 1/2 -0.0186(1) 0.6172(1) 22(1) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[a2811 + b2822 + 92833 + ab(cos 10812 + 41de 0813 + bclcos (1)323]- Table 66. Fractional Atomic Coordinates and Bag Values for NaCuTe with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Te 1/4 1/4 0.7224(1) 124(2) Cu -1/4 1/4 12 246(4) Na 1/4 1/4 1.172(1) 1.9(1) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4I3)[aZB11 + b2822 + 02833 + ab(cos 0812 + ac(cos B)Bta + bc(cos (1)323]- 182 3. Results and Discussion 3.1. Synthesis The synthesis of two novel layer compounds of K2CusTe5 and NaCuTe has been achieved using polytelluride fluxes as solvents and reagents at 300400 0C as shown in eq 1. nA2Te + Cu + 6Te -------- > K2CU5Te5 or NaCuTe eq 1. where A is K or Na metal and n is 3 or 5, respectively. They form complete melts at the temperature employed (350 °C for (I) and 450 0C for (ID). The equation given above is not balanced and only shows the reactants used (left- hand side) and product obtained (right-hand side). After reaction, excess polytelluride flux was removed from the product with extra caution by washing with degassed DMF. During washing, we have often seen decomposition of the polytelluride flux into elemental tellurium in the presence of oxygen or water. Therefore, isolation has to be done carefully under an inert atmosphere using degassed DMF. Both compounds are stable in the air and moisture for several days. The preparation of K2CU5Te5 and NaCuTe was carried on without difficulty. They were readily obtained from the K2Tex and Na2Tex fluxes as pure phases. However, K2CU5Te5 caused some difficulties in obtaining pure another ternary KlCu/l'e compound, K4CU3T611 (see the synthesis section in chapter 6 for detail), since the former was a competing phase in the preparation of the latter. When we used a polytelluride flux with shorter chains (e.g. K2T63) the known compound, KCU3Te2102 was obtained at 350 0C. The NaCuTe 183 phase, which had been prepared at 800 0C by direct combination of the elements", was isolated with every reactant ratio Na2Te/Curr e of 1/1/8 to 5/1 l8 at lower temperature (400 0C). Among them, higher Na2Te ratio of 5/1/8 gave better yield and crystallinity. The rest of the ratios showed relatively poor crystallinity and had elemental tellurium as coexisting phase. 3.2. Description of Structures 3.2.1. Structure of K2Cu5Tes (I) The structure of K2CU5Te5 is anisotropic and is shown in Figure 38. Anionic [CusTe5]n2"' corrugated layers alternate with charge compensating K+ ions. The [CusTesanII' layers are formed by edge-sharing (fused) rhombic [CU2Te2J units to form a distorted anti-PbO like structure as shown in Figure 39. The structure of K2CU5Te5 can be viewed as deriving from the anti-PbO like structure of CuTe104 into which K+ ions are reductively inserted. The two- dimensional layered structure of CuTe is shown in Figure 40. It contains unusual infinite straight Te chains along the a-axis with a Te-Te bond distance of 3.10 A. In this structure Cu atoms occupy the O sites and Te occupy the Pb sites of the PbO structure. Based on the formula K2CU5Te5 and on the assumption that the oxidation state on each Te atom is -2, the formal oxidation state on copper is +1 (two atoms) and +2 (three atoms). However, the coexistence of Cu2+ and T62' is thermodynamically unstable with respect to electron transfer from the reducing TeZ' to the oxidizing Cu2+. If one then considers all copper atoms in the +1 oxidation state, it follows that the average oxidation state of Te is -1.4. This electron deficiency (holes) on Te can be either delocalized in the valence band or localized as ditelluride via a structural distortion or both. This situation is reminiscent of that found in CuSGZ, a known 184 00.90.2092 .0 :00 :5 05 .0 0.550 05.000. 05 552500502 awhmo .mm 059”. 20 30.00 O O .. .. .......... 2.2.0.. . ...:7...\..//\\.../ .. ......MIMM. K... «t ‘61.! , [..2..\2.’2»~V\22’<\4 . ‘62. I. l. a .3 a 0 VJ M... 38 .222... €7.22. 0 . 0KO 0m. :69 D 869.. 8.2.2” O D \\O 0 0 W cm lly O 185 [G1 5 TC: 1:“- - -. .- gunman...» .‘r-w'wln" I . ...-......qu 22.222222"... .Imlslslnlt. [CuTe] " Figure 39. Structural relationship of the [CU5T65]n2"' layer to those of CuTe (above) and [CuTe]n"' (below) as a function of oxidation state. 186 Cu U. / \ W T Figure 40. Two axial views of the layered structure of CuTe: (A) a-axial view and (B) b-axial view. 187 metallic mixed-valence compound formulated as CU2+CU2+(822')(32') or as CU3+(S22')(S')1°5. ln K2CU5Te5 there are three crystallographically independent copper atoms which possess a distorted tetrahedral geometry. The average Cu-Te distances for Cu(1), Cu(2) and Cu(3) are in the range of 269(3) A, 264(7) A and 2.58(4)A respectively, and compare favorably with those found in CuTe104. However, the Cu(2)-Te(1) and Cu(3)-Te(3) distances (along the crystallographic a-axis) are short at 2.582(1) A and 2.540(2) A, respectively, suggesting that the hole might be constrained but delocalized on rows made from these atoms parallel to the a—axis. Based on the +1 oxidation state for Cu, there are three electron vacancies per five Te. As a result, a structural distortion from the ideal anti-PbO structure occurs in which two vicinal Te(2) atoms move closer together along the c-axis to form a bond. Structural distortions in the electronegative element network as a response to changes in the electronic structure of the electropositive metal are at the core of the Zintl concept91 and are found in other ternary chalcogenide and pnictide systems106. There are three crystallographically distinct Te atoms in the asymmetric unit of K2CU5Te5. Among them, Te(2) atoms form Te(2)-Te(2) bonds of 2.960(1)A by localizing 2 holes. In the absence of an additional distortion we can expect the remaining electron vacancy to be delocalized on the valence band, consisting mainly of p orbitals of the remaining three Te atoms. This gives rise to a partially empty band. The Te(2)-Te(2) bond is longer than the normal Te-Te single bonds found in known polytelluride compounds107 by ~0.20 A due to the multidentate nature of the Te(2)-Te(2) ligand. Each Te(2) atom binds four Cu atoms which pull a significant fraction of the electron density from the Te-Te bond. The next shortest Te-Te distance is in the non-bonding range of 3.667(1) A. The square pyramidal geometry around Te(1) and Te(3) 188 and their comparable average Cu-Te distances (261(3) A for Te(1) and 263(9) A for Te(3)) suggest indistinguishable oxidation states on Te(1) and Te(3) even though they are crystallographically distinct. K2CU5Te5 can be reasonably formulated as K2+CU5+(Te22')Te31-67'. The Cu atoms in the [Cu5Te5]n2"' layer make short Cu-Cu contacts with each other in the range of 2.687(2) A-2.909(2) A. Short Cu-Cu contacts are not uncommon in the copper chalcogenide compounds. Selected bond distances and angles are given in Table 67. Charge balancing K+ ions are distributed between the layers, making ionic interactions with 10 Te atoms. The shortest K-Te distance is 3.561(2) A. 189 Table 67. Selected Bond Distances (A) and Angles (deg) in K2CusTe5 with Standard Deviations in Parentheses Cu(1)-Te(1 ) 2.644(2) Cu(1)-Te(1)-Cu(2) 89. 10(5) Cu(2)-Te(1 ) 2.637(1) Cu(1)-Te(1)-Cu(2) 6763(4) (x2) Cu(2)-Te(1) 2.582(1) (x2) Cu(2)-Te(1)-Cu(2) 61 .96(4) (x2) Cu(2)-Te(1)-Cu(2) 106.24(6) Cu(1)-Te(2) 2.701 (1 ) (x2) Cu(1)-Te(2)-Cu(1) 9974(6) Cu(2)-Te(2) 2.760(2) Cu(1)-Te(2)-Cu(2) 6435(4) (x2) Cu(3)-Te(2) 2.625(1 ) Cu(1 )-Te(2)-Cu(3) 6429(3) (x2) Cu(2)-Te(2)-Cu(3) 95.51 (4) Te(2)-Te(2) 2.960(1 ) Te(2)-Te(2)-Cu(1) 9857(3) (x2) Te(2)-Te(2)-Cu(2) 151 .19(3) Te(2)-Te(2)-Cu(3) 5568(2) Cu(1)-Te(3) 2.713(1) Cu(1)-Te(3)-Cu(3) 6522(3) (x4) Cu(3)-Te(3) 2.540(2) (x2) Cu(3)-Te(3)-Cu(3) 108.79(7) (x2) Cu(1)-Cu(3) 2.834(2) Te(1 )-Cu(1)-Te(2) 1 1390(4) (x2) Cu(1)-Cu(2) 2.909(2) Te(1)-Cu(1)-Te(3) 108.38(4) Cu(2)-Cu(2) 2.687(2) Te(2)-Cu(1)-Te(2) 9974(6) Cu(2)-Cu(2) 2.687(2) Te(2)-Cu(1)-Te(3) 1 1037(4) (x2) Te(1 )-Cu(2)-Te(1) 118.04(4) (x2) Cu(1)-Te (mean) 269(3) Te(1)-Cu(2)-Te(1) 106.24(6) Cu(2)-Te (mean) 264(7) Te(1)-Cu(2)-Te(2) 8586(4) Cu(3)-Te (mean) 258(4) Te(1)-Cu(2)-Te(2) 113.96(4) (x2) Te(2)-Cu(3)-Te(2) 6863(5) Te(2)-Cu(3)-Te(3) 118.74(2) (x4) Te(3)-Cu(3)-Te(3) 1 0879(7) K-Te(1 ) K-Te(1 ) 3.611(2) (x2) 3.712(2) (x2) K-Te(2) K-Te(3) 3.630(2) (x2) 3.561 (2) (x4) 190 3.2.2. Structure of NaCuTe (II) The structure of NaCuTe has been known for a while.74 However, its poor characterization of the structure and quite important structural relationship with the KZCusTes and CuTe renders us to redetermine its structural details with high accuracy. NaCuTe possesses an ideal anti-PbO-type layer and isostructural to NaCuSe. The structure is shown in Figure 41. The anionic [CuTeJn ' layer (shown in Figure 39) is made of edge-sharing rhombic [CuzTeZ] units. The structure of NaCuTe can also be viewed as being derived from the distorted anti-PbO-type structure of CuTe into which Na+ ions are reductively inserted. The Cu-Te bond distance of 2.709(1)A compares well with that of NaCuaTe2101. The tellurium atoms have 114-type coordination with square pyramidal geometry. Selected bond distances and angles are given in Table 68. Charge balancing Na+ atoms are distributed between the layer, making ionic interactions with Te atoms. The average Na-Te distance is 3.197(3)A. The geometry of the 5 coordinate Na+ ions is square pyramidal. Table 68. Selected Bond Distances (A) and Angles (deg) in NaCuTe with Standard Deviations in Parentheses Te-Cu 2.7086(6) Cu-Te-Cu 6995(2) (x4) Cu-Te-Cu 108.31 (3) (x2) Te-Na 3.203(7) Te-Cu-Te 108.31 (3) (x2) Flgure 41. ORTEP representation of the layered structure 01‘ NaCuTe. 192 3.3. Structural Relationship of the [CusTesln'M' layers in (I) to those of [CuTelnfl' In (II) and CuTe The layered compound CuTe, which is formally better described as (Cu+)n(Ten"') rather than (Cu2+)(Te2‘), contains unusual, straight Ten 'chains. The long Te-Te bonds at 3.10 A are considered half bonds and are subject to reduction by accepting electrons from the electropositive elements. Thus, the structure of the [CusTeshZN' layer can be rationalized as being derived from that of CuTe by addition of 0.4 electrons. This reduction breaks some of the Te- Te bonds in CuTe. One electron per CuTe unit is required to reduce all Te-Te bonds of the infinite Tenn“ chain as is found in NaCuTe. The Te-Te bonds of the infinite, straight Tenn' chains are broken by accepting electrons from the electropositive Na atoms to give the ideal anti-PbO-type layered structure of [CuTe]nn'. Thus, K2CU5Te5 can be considered as a true intermediate oxidation state compound between completely oxidized CuTe104 and completely reduced NaCuTe as illustrated in Figure 39. The known layered compound KCuTe73 can also be considered as being derived from the CuTe framework by reduction with 1 electron. Surprisingly, the structure of KCuTe differs from that of NaCuTe. It forms the boron-nitride honeycomb layer structure shown in Figure 42. The structure change of the anionic framework from Na to K is purely an electrostatic effect. The larger K+ ion requires a larger area and volume than does the Na+ ions. This destabilizes the PbO structure which requires that the closest K+---K+ distances be the same as the Nat-“Na+ distances. For the same number of formula units of [CuTe]', the boron-nitride structure type covers more area than the PbO structure type. Thus, enough space is created to accommodate the K+ ions. Based on these observations, it is possible to predict the CsCuTe Figure 42. ORTEP representation of (A) the [CuTelnn' layer and (B) the unit cell of KCuTe. 194 structure which is not identified yet. Since ionic radius of Cs+ is much larger than that of K+ ion, much larger areas and volume would be required for the 05*” ion. This would destabilize the boron-nitride framework of [CuTelnn' and form one-dimensional structure which is similar to that of KCuS72. 3.4. Charge Transport Properties of chusTes (I) Based on the band filling arguments advanced above, KZCusTes is expected to be a p-type metallic conductor. Charge transport measurements over the temperature range 5-300 K on the single crystals of chusTes along the needle axis show that the resistivity first decreases linearly with decreasing temperature and at low temperatures levels off to a constant value (so called residual resistance due to scattering by impurities). The data are shown in Figure 43. The resistivity increases from 3.1x10'6 0cm at 5 K to 6.7x10'5 0cm at room temperature. The conductivity values of K20U5Te5 are among the highest for known copper chalcogenide compounds. The residual resistance ratio (rn/ro), which is often used to determine the purity of the metal, is 21.3 in K20U5Te5. In some of high purity metals such as copper and aluminum, the ratio can be ~200. The temperature dependence of the thermoelectric power (Seebeck coefficient) shows a very small positive value of.1~3 pV/K in the temperature range of 25~300 K as shown in Figure 44. The small and almost constant Seebeck coefficient indicate that chusTes is a p-type metal. The small increase of the thermopower at low temperature (onset at 50K) is probably due to scattering by impurities. 195 .mokmnowx so .995 29% m .2 2293.5. .o 8:25. a mu Emu €8.33 >=>=m_mo. coca Son. .9 93mm 6: oSfiEanh com 0mm com om, cop on o .IJIll4llJ, q — d u «llAl dldl. — .11ile—qulll1ll _ll-_i.l.d ._ -_ --AIIJ l— l —---4ll-4|t_lll o (“JO-U”) AllAllSlsaH .mohmaomx _o .3050 06% m .2 San Q35 .038 058.0252. QBSKEQoBmtg .3 2:9”. 0: 9398th con omN com omp cop cm 0 q q q q — d u d 4 — q d « d — fl a fi — a a q a — a d a «I; L 196 ()(//\11) laModouueq J. 197 3.5. Magnetic Susceptibility of chusTes (i) Magnetic susceptibility data for chusTes in the temperature range 2- 300 K at 5 k6 are shown in Figure 45. When plotted as x" vs temperature, the data above 30 K show temperature-independent paramagnetism while below 30 K the data follow a Curie-Weiss Law. This behavior is characteristic of metals (Pauli paramagnetism) containing a low concentration of paramagnetic impurities. In the free electron gas metal, the Pauli spin magnetization of the conduction electrons is given by eq 2.108 3Np. ZKBTp Where (I is Bohr magneton (0.921x10'20 erg/0e), B is the magnetic field intensity, K3 is the Boltzman constant, N is number of carriers, and T1: is the Fermi temperature. Landau has shown that for free electrons the magnetic field induces a diamagnetic moment equal to -1/3 of the paramagnetic moment. Thus, the total magnetization of a free electron gas is given by eq 3. M= B=—— eq3. 198 H m> 3.5295 ..Ex ”tome. .mohmanuwx oc___m.mao>.oa .2 San 995353 £33083 2.2.8:. o.29an£.o.nm=m> .mv 2:98 3.: com com oo— o . . . s c h B I B B a I I I I III Imam-EB man a L m ask a 3 on S o. o .. . . p b . . p c IrN 00009 W. a 0000 IF \I . . . m _..: e a . .N m ..eepx. m (low/nuts) wx 199 EF is the Fermi energy which may be estimated from the eq 4 where m is the mass of the electron, N is the number of carriers, and the V is the molar volume. Thus NN is the conduction electron concentration. h2 anN 213 Er = X —_ eq 4. 2m V For K20U5Te5, we expect one carrier per formula unit. Knowing the molar volume (184.4 cm3), NN is then 3.265x103?1 cm'3 and the Fermi energy is calculated as E(:=1.286x10'12 erg. The predicted magnetic susceptibility is then xp=39.7x10'5 emu/moi. The temperature independent Pauli paramagnetism of the sample (xp=25.6x10'5 emu/moi) has been obtained by subtracting the Curie-Weiss portion of the paramagnetism from the measured magnetic susceptibility. This value compares reasonably well with the predicted value of 39.7x10'5 emu/moi. CHAPTER 6 Synthesls and Characterization of K4Cu3Te11 and Cs30u3Te1o: Novel Solid State Chaloogenide Compounds with a Dodecahedral Cluster as a Building Block 1. Introduction in the last decade there have been great efforts to search for new materials with interesting electrical, optical and catalytic properties. We have been pursuing this goal by exploring new ternary solid metal chalcogenide compounds using molten salt synthetic methods. Particularly, the use of alkali metal polychalcogenide fluxes as solvents and reagents has been quite successful in synthesizing novel structural types of (poly)sulfides and (poly)selenides of various transition-metals at intermediate temperatures (150 < T < 500 °C)50.53 (see previous chapters 1-4). We also have initiated investigations into polytelluride melts and have discovered several new compounds with unusual structures (see chapter 5). A new mixed-valence compound K2CusTe5 discovered in our laboratory was reported earlier.54(b) Metal tellurides have rarely been investigated compared to the corresponding metal sulfides and metal selenide36t7v79ldi-100. In this chapter, two novel Cufl' e cluster compounds, K40U3Te11 and CseCuaTem, with complicated structures will be illustrated. Prior to our work in the AlCu/T e (A=alkali metal) system there 200 201 were only four known phases, NaCuaTe2101- KCuaTe2102, NaCuTe (see page 190), and KCuTe74, all of which contain only mono-telluride ligands. 2. Experimental Section 2.1 Reagents Chemicals were used as obtained: copper powder, -325 mesh, 99.95% purity, Cerac, Milwaukee, WI; tellurium powder, -100 mesh, 99.95% purity, Aldrich Chemical Co., Milwaukee, WI; potassium metal, analytical reagent, Mallinckrodt lnc., Paris, KY; cesium metal, 99.98% purity, AESAFI, Johnson Matthey, Seabrook, NH. 2.2. Physical Measurements Magnetic susceptibility measurements for CsaCu8Te1o over the temperature range from 5 K to 300 K at 5 k6 were performed on a MPMS Quantum Design SQUID magnetometer. Single crystals of CsaCugTew were manually selected for measurements. They were used without grinding as random oriented single crystals. The data were corrected for diamagnetic contributions of the sample holder. To obtain molar susceptibility the corrections for ion-core diamagnetic contributions from atomic constituents were made using the values tabulated by Mulay103. The magnetization of CS3CueTe1o was examined at 5 K as a function of applied field from 500 G to 7 k6 and was found to vary linearly with the applied field. Four-probe dc resistivity data for K4CU3Te11 and thermoelectric power data for 0330U3Te1o over the temperature range 5 K to 300 K were provided by 202 Prof. C. R. Kannewurf (Northwestern University). A computer automated measurement system was employed to obtain thermopower and resistivity data with both the current and thermal gradient applied along the needle axis of K40U3Te11 and (010) plane of CsaCusTem. For all measurements electrode connections to the small single crystals were made with the use of 25 and 60 um gold wires and gold bonding paste. Quantitative microprobe analysis of the compounds were performed on a Jeol 35CF scanning electron microscope equipped with Tracor Nothern TN 5500 X-ray microanalysis attachment. Single crystals of each sample were carefully picked and mounted on an aluminum stub using conducting silver paint to help dissipate charges that developed on the sample surface during measurements. Energy Dispersive Spectra (EDS) were obtained using the following experimental set-up: X-ray detector position : 55 mm Working distance : 39 mm Accelerating voltage :20 KV Take-off angle : 27 deg Beam current : 200 picoamps Accumulation time : 100 seconds Window : Be A standardless quantitative analysis (SQ) program was used to analyze the X-ray spectra obtained. Since the Cu ratio is always overestimated because of the contribution of system Cu peaks, a correction factor (x 0.73), determined by calibrating with known K/Cufl e compounds, was used to better evaluate the Cu percentage. 203 2.3. Synthesis Chemicals were measured and loaded in Pyrex tubes under a dry nitrogen atmosphere in a Vacuum Atmospheres Dri-Lab glovebox. Potassium monotelluride (K2Te) and sodium monotelluride (CszTe) were prepared in liquid ammonia from alkali metal and elemental tellurium in a 2:1 ratio. Tetrapotassium bls(p4-dltelluro)tris(p3-ditelluro)(u4-telluro) octacuprate(l), K4Cu3Te11 (I) 0.309 g (1.5 mmol) of KzTe, 0.064 g (1.0 mmol) of Cu and 0.765 g (6.0 mmol) of Te were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 350 °C for 3 days and cooled slowly to 100 0C at a rate of 2 °CIhr. Black needle-like crystals, sometimes with small contamination of elemental tellurium, were obtained by removing excess molten potassium polytellurides (K2Tex) with DMF under a N2 atmosphere (yield: 57% based on copper used). A quantitative microprobe analysis performed on a large number of crystals with the EDS/SEM system gave an average composition of KCuz,1Te2,7. Tricesium bls(p4-dltelluro)bls(ps-dltelluro)bls(p4-telluro) octacuprate(l,ll), CsoCuaTe1o(ll) 0.197 g (0.5 mmol) of CszTe, 0.032 g (0.5 mmol) of Cu and 0.383 g (3.0 mmol) of Te were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum («40‘3 torr). The tube was placed in a computer-controlled furnace and heated at 450 0C for 4 days and cooled slowly to 100 0C at a rate of 2 oC/hr and then to 50 0C at a rate of 10 0CM. Small amounts of black, plate-like crystals were obtained as a minor phase with large amounts of powder material including elemental tellurium. The matrix, a hard mass, does not dissolve easily in DMF and water or any 204 other common organic solvent. Thus, we have crushed the matrix to manually isolate single crystals. A quantitative microprobe analysis performed on a large number of single crystals with the EDS/SEM system gave an average composition of CsCuz,5Te3,3. 2.4. x-ray Crystallographic Studies Both compounds were examined by X-ray powder diffraction for the purpose of phase characterization and identification. The d-spacings for each compound were obtained from the powder pattern recorded on a Phillips XRG- 3000 computer-controlled powder diffractometer, operating at 40KV, 35 mA. Graphite monochromated Cu radiation was used. To verify product homogeneity, the d-spacings observed for the bulk materials of K40U3Te11 were compared, and found to be in accord, with those calculated from the single crystal X-ray structure analysis data. The d-spacings observed for single crystals of CS3CU3Te1o, which were manually selected, are also in accord with those calculated from the single crystal X-ray structure analysis data. The calculation of d-spacings was performed using the POWD10 program56. The results are summarized in Tables 69 and 70. 205 Table 69. Calculated and Observed X-ray Powder Diffraction Pattern of K40U6Te11 W" at H K L MM) 60mm) I/Imax(obs.) -2 0 1 11.97 12.03 21.3 2 0 0 9.93 10.0 6.7 -4 0 1 5.76 5.62 55.5 4 0 3 5.36 5.42 7.3 -3 1 3 4.50 4.53 11.3 4 0 1 4.09 4.11 6.3 .4 0 5 3.69 3.71 16.3 -2 0 5 3.52 3.54 21.3 6 0 0 3.31 3.33 36.7 -3 1 5 3.23 3.25 26.0 0 2 2 3.11 3.06 35.0 5 1 1(1 1 4) 3.04 3.06 44.7 4 2 1 2.94 2.95 16.7 6 0 1 2.91 2.92 35.5 -8 0 5(-4 2 3) 2.63 2.69 100 4 0 3 2.34 2.65 20.9 2 2 2 2.79 2.60 12.7 -8 0 6 2.69 2.70 16.3 6 0 0 2.463 2.485 6.9 -10 0 7 2.219 2.226 36.7 8 0 2(6 2 2) 2.044 2.050 65.2 Table 69. (cont'd) H K L deA) dobs(A) lllmax(obs.) -5 3 1 2.031 2.032 23.1 6 2 0(-6 2 7) 2.007 2.004 24.5 -2 2 7 1.960 1.965 37.3 -7 1 9(-10 2 5) 1.963 1.963 24.5 -9 1 9 1.9130 1.9195 23.1 1 3 4(-12 0 6) 1.6699 1.3666 54.1 -2 0 9 1.8496 1.6541 26.0 -5 3 6 1.3261 1.6330 16.3 -10 0 10 1.7937 1.7972 10.0 «4 0 10 1.7612 1.7643 14.2 0 4 0 1.7053 1.7244 6.6 -14 0 4 1.6752 1.6766 15.5 -2 0 10 1.6511 1.6556 6.0 207 Table 70. Calculated and Observed X-ray Powder Diffraction Pattern of C63Cu8Te1o H K L await) 60mm l/Imax(obs.) 0 2 0 12.07 12.40 3.3 0 4 0 6.03 6.01 24.2 1 4 1 3.32 3.88 21.7 0 0 2 3.46 3.50 20.1 1 61(0 71) 3.12 3.12 100 (1 1 2) 3.08 (2 4 0) 3.04 2 3 1 2.92 2.94 29.4 3 0 1 2.226 2.236 17.9 2 6 2(111 0) 2.106 2.111 41.4 1 9 2 2.032 2.027 36.2 010 2 1.962 1.964 33.2 1 6 3(2 1 3) 1.9233 1.9350 27.6 3 5 2 1.6050 1.6094 16.5 0 0 4 1.7337 1.7359 41.5 3 9 2(4 0 2) 1.5754 1.5773 21.7 208 The X-ray single crystal data of K4Cu8Te11 and CsaCuaTem were collected on a Rigaku AFC68 diffractometer with graphite monochromated Mo-Ka radiation using the (”'26 scan mode. Accurate unit cell parameters for both compounds were obtained from the least-squares refinement of the 20, 00, x, and 4) values of 20-25 machine-centered reflections. The stability of the experimental setup and crystal integrity were monitored by measuring three standard reflections periodically (every 100 reflections) during data collection. The intensities did not show any appreciable decay. Two absorption corrections were applied to the data of K40U3Te11 and CsacuaTem: an empirical absorption correction based on 1p scans for three reflections followed by a DIFABS57 correction. The structures were solved by direct methods with SHELXS-8658 program and were refined with the TEXSANGo package of crystallographic programs. All calculations were performed on a VAXstation 3100 computer. All atoms were refined anisotr0pically. The complete data collection parameters and details of the structure solution and refinement for (I) and (II) are given in Table 71. The final atomic coordinates, temperature factors and their estimated standard deviations are given in Tables 72 and 73. 209 Table 71. Summary of Crystallographic Data for K4CU3Te11 and CS3CU3T610 mild I II Formula K40U3Te1 1 CsaCuaTe1 0 Formula weight 2068.36 2183.08 space group C2Im Immm ... a(A) 24.066(3) 7.053(2) F 5(A) 6.621(6) 24.159(3) c (A) 16.461(3) 6.935(3) 3 (deg) 90.0 90.0 6 (deg) 124.45(1) 90.0 i 7 (deg) 90.0 90.0 L Vol (A3), 2 2501(3), 4 1182(1), 2 Temperature (°C) 23 23 Crystal size (mm) 0.60x0.06x0.06 0.39x0.18x0.08 Radiation Mo-Ka Mo-Ka p. (Mo-Ka, cm°1) 199.8 237.4 Dcalc (g/cm3) 5.49 6.14 20".” (deg) 50 60 Scan method (0’20 (1)/29 No. of data collected 2535 1014 No. of unique data 2480 1014 No. of data used 2029 839 (F02>301F02)) No. of atoms 17 7 No. of variables 1 21 36 Phasing technique Direct methods Direct methods Final R/RW 2513.7 3.2/5.0 Max. shift/esd 0.00 0.00 (last cycle) Extinction coefficient 3.50x10'7 205x10'7 210 Table 72. Fractional Atomic Coordinates and ng Values for K4CU3Te11 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Te(1) 0.03755(4) 0 0.13004(5) 1 . 12(3) Te(2) 0.09026(4) 0 0.31031 (5) 1 . 14(3) Te(3) -0.06544(4) -1 I2 -0. 01 854(5) 1 . 14(3) Te(4) -0.09455(3) -0. 29501 (9) 0.181 19(3) 1 .36(2) Te(5) 0.07121(4) -1/2 0.44196(5) 145(3) Te(6) 0.19740(4) -1/2 0.46792(5) 119(3) Te(7) 0.22661 (3) 0.29496(9) 0.27474(4) 144(2) Te(8) -0.04266(4) 0 0.40721(5) 147(3) Te(9) -0. 1 8848(4) 0 -0. 05765(5) 1 30(3) CU( 1) 0.00740(5) -0. 2060(2) 0.33673(7) 1 .81 (4) Cu(2) 0.20463(6) 0.1 924(2) 0.38766(7) 1 .90(4) CU(3) -0.07933(6) -0.1 939(2) 0.05821 (7) 1 .68(4) Cu(4) 0.1 2886(5) -O. 1 932(2) 0.1 1857(7) 1 .65(4) K(1) 0.0640(1) -1I2 0.2222(2) 1.9(1) K(2) -0.1 274(2) 1/2 0.3299(2) 2. 7(1 ) K(3) 0.3541(1) 12 0.4743(2) 2.1(1) K(4) -0.2499 0 0.0867(2) 3. 8(2) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4I3)[aZB11 + b2822 + 02833 + ab(cos y)B12 + ac(cos (3)813 + bc(cos (1)823]. 211 Table 73. Fractional Atomic Coordinates and Beq Values for CSsCUaTem with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 Cs(1) 1/2 0.22267(6) 0 296(6) Cs(2) 1/2 0 12 096(5) ? Te(1) 172 0.05306(4) 0 099(4) f Te(2) 0 0 0.2962(1) 102(4) Te(3) 0.3026(1) 0.14514(3) 12 137(3) . Te(4) 0 0.16360(5) 0 127(4) L Cu(1) 0.1946(2) 0.09123(5) 1.1956(1) 1.62(4) a B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4I3)[a2811 + [32322 + 62833 + ab(cos y)B12 + ac(cos 19313 + bO(COS 00323]- 212 3. Results and Discussion 3.1. Synthesis The synthesis of two novel cluster-containing compounds, K4Cu8Te11 and CSaCUaTem, has been achieved using polytelluride fluxes as solvents and reagents as shown in eq 1. nA2Te + Cu + 6Te -------- > K40U3Te11 or CS3Cu8Te1o eq 1. where A is K or Cs metal and n is 1.5 or 1, respectively. They form complete melts at the temperatures employed (350 °C for (l) and 450 0C for (Il)). The equation given above is not balanced and only shows the reactants used (left- hand side) and product obtained (right-hand side). After reaction, excess polytelluride flux was removed from the product with extra caution by washing with degassed DMF. The polytelluride flux may decompose to elemental tellurium if water or oxygen are present. Therefore, isolation has to be done carefully under a inert atmosphere with degassed dry DMF. Both compounds are stable in the air and moisture for several days. K40U3Te11 was prepared at 350 0C using the reactant ratio K2Te/Cu/T e of 111 I8 The product contained KZCusTes (see page 183) and some amounts of tellurium as coexisting phases, as judged by X-ray powder diffraction pattern. In order to suppress the KzCusTe5 and elemental tellurium phase, we made several variations in the reaction conditions. First, upon changing the Cu ratio from 1 to 2.5 in the K2Te/CulT e system at 350 00, we obtained more or less the same results with elemental tellurium as the major phase. However, the 1/2/8 ratio yielded slightly more K4CU3Te11. Then we increased the K2Te ratio from 1 213 to 4 to optimize the metal to flux ratio. The 4/218 ratio gave KCuaTe298 with tellurium contamination while the 212/8 ratio gave only KzCusTes. Only the 3/2/8 ratio gave K4003Te11 phase as the major product at 350 °C. This result seemed reasonable because the KICu ratio of K40uaTe11 is slightly higher than that of KZCusTes. However, the product was still contaminated with minor K2CusTe5 and Te phases. An additional variation in the reaction condition was made by increasing Te from 3/2/8 to 3/2/12 based on the fact that K40u8Te11 has a higher Te/Cu ratio than K2CU5T65. Indeed, the 3/2/12 ratio yielded K4CueTe11 as the major product with complete suppression of K2CU5Te5. However, it is still contaminated with small amounts of tellurium. A parallel approach, made by increasing the temperature (400 0C and 450 °C) with the reaction ratios mentioned above, was unsuccessful. At these high temperatures K4CU3Te11 was suppressed while the competing KZCusTe5 phase is stabilized. This is consistent with the known property of decomposition of polychalcogenides to monochalcogenides at high temperature. In the CszTe/Cu/T e system, we have seen at least two new phases. The reaction has been carried out using reactant ratios CszTeICu/l' e of 1“ I8 to 41118 at 330 °C. The 1/1/8 to 2/1/8 ratios gave a few large single crystals of CsaCuaTem. The majority of the product was elemental tellurium. The ratio of 3/1/8 to 4/1/8 gave another new ternary compound which has approximate composition of CsCuTe4 given by EDS/SEM quantitative analysis. However, so far, we have been able to isolate single crystals of this compound. Its thin, hair- like morphology was identified using scanning electron microscope. The yield of CsacueTem, a minor phase in the preliminary trial, was improved slightly by changing reaction conditions. We increased the Cu ratio from mm to 1/3/8 at 380 °C and then reduced the Te ratio from 1/1/8 to 111 I4 and increase the temperature (450 °C). The CsaCuaTem phase isolated from fluxes with these 214 ratios was still only a minor phase, but the crystal sizes were as large as 3mm. Only the 1/1 I6 ratio gave a better yield but still more work is required to improve the overall yield and homogeneity. Isolation of the homogeneous CSQCU3T610 phase and single crystals of the new CsCuTe4 phase is currently under investigation. 3.2. Description of Structures 3.2.1. Structure of K4Cu3Te11 (I) The structure of K40U3Te11 is a unique three-dimensional CuIT e framework with large tunnels running parallel to the crystallographic b-axis, as shown in Figure 46. The tunnels are filled with K+ ions. The framework contains tetrahedral Cu+ centers bonded to Te2' and Te22' ligands. The formula unit can be represented as K4CU3(Te2)5Te. The structure is somewhat complicated with its three-dimensionality, but is tailored from fused and linked recognizable Cufl’e clusters. The basic building block of this framework is the remarkable pentagonal dodecahedral cluster, Cua(Te2)5, as shown in Figure 47(A). A remarkable feature of the dodecahedral Cua(Te2)6 cluster is the encapsulation of a K+ ion in its center. This dodecahedral cluster is made of fused CuzTea pentagonal planar five-membered rings each with one ditelluride edge. These Cua(Te2)5 clusters contain three mutually perpendicular sets of ditelluride units. Two dodecahedral Cu8(Te2)6 clusters share one Te-Te edge to form a "Siamese twin" type double cluster shown in Figure 47(B). These double clusters then share opposite Te-Te edges to form a straight one-dimensional column with oval cross-section, as shown in Figure 47(0). The Cu atoms in these columns are bridged by quadruply bonded p4-Te2' ions above and below the columns. This results in another unusual Cuff e cluster as shown in Figure 215 Figure 46. ORTEP representation of the unit cell of K4CuaTe11 viewed down [010] direction. 216 Figure 47. ORTEP representation and labeling scheme of (A) the [KCuaTe12] dodecahedral cluster, (B) two edge-shared [KCuaTe12] units with capping Te2' ions, (C) a one-dimensional column of edge-shared double clusters, and (D) the empty CUaTea cluster. 217 NV 239". L. Ch. 5.. . \/ \/ \/.w 5:- .. 2f .\ IL; )1»; \ IL. . .V3 3.6 \ w _ 21.1.1, _Llnlet 2er7 . A/ \ A/.\v A, \v.. a .... .WV/JWWVAWMWV/Q. . .2. 3.6 / . . 4.... II rat a e . ‘ 3 ...III,. Ill .11.... 5.. . \ 7/ \\ 7/ \\ 7/ \t E: E. .2 .3 218 47(D). This smaller cluster is made of four pentagonal CuzTea five-membered rings and eight puckered CuzTez four-membered rings. The inside of this CuaTeg cluster is completely empty with dimensions of 4062/) x 6.821A x 8068A corresponding to the distances of Te(1)-Te(1), Te(3)-Te(3) and Te(9)- Te(9), respectively. The one-dimensional columns of Cu3(Te2)3 clusters are then assembled side by side through intercolumn Te-->Cu bonding interactions to form CuIT e layers as shown in Figure 48. The Te(6) atoms that participate in these interactions are from one of the non-edge-shared ditelluride units (is. Te(5)-Te(6)) in the dodecahedral Cua(Tez)5 cluster. The CuITe layers are then connected to each other via bridging ditellurides (i.e. Te(8)-Te(8)) that act as pillars between the layers, producing the large channels as shown in Figure 46. The cross section of the channels is a 24 membered ring which is roughly rectangular-shaped; its short dimension is 4.581 (1)/1. The geometry around the Cu atoms is distorted tetrahedral. The average Cu-Te distance is 263(4)A which is in the normal range of Cu-Te distances. Short Cu-Cu contacts are also observed in this compound, ranging from 2.625(3)A for Cu(2)-Cu(2) to 2.610(4)A for Cu(1)-Cu(1). The coordination environments of Te atoms vary. The p4-Te(9) atoms have a square pyramidal geometry. The tellurium atoms (T e(1), Te(2), Te(3), Te(6)) of ditellurides are bonded to four Cu atoms and one Te atom with a square pyramidal geometry, while the tellurium atoms (T e(4), Te(5), Te(7) and Te(8)) of ditellurides are bonded to two Cu atoms and one Te atom with a trigonal pyramidal geometry. Higher coordination numbers of Te atoms are not uncommon in the binary or ternary compounds, some examples of which include KCU482Te (see page 255), NaCuaTe2101, KCuaTe2102, and TlCuaTe2109. The average Te-Te distance of ditellurides is normal at 281(1) 14.103 Selected bond distances and angles are given Table 74. .K)‘ :2)‘ 1°. >11» - :: lo .1°> in . x 7‘ Figure 48. ORTEP representation of (A) sideview and (B) top view of one CuITe layer. The Cu3(Teg)5 and CusTes clusters are shaded for emphasis. Table 74. Selected Bond Distances (A) and Angles (deg) in K4CuaTe11 with 220 Standard Deviations in Parentheses Te(1 )-Te(2) Te(1 )-Cu(3) Te(1 )-Cu(4) Te(2)-Cu(1 ) Te(2)-Cu(2) Te(4)-Cu(3) Te(3)-Te(3) Te(3)-Cu(3) Te(3)-Cu(4) Te(4)-Te(4) Te(4)-Cu(1 ) Te(5)-Te(6) Te(5)-Cu(1 ) ¥ 2.312(1) 2.665(2) (x2) 2.673(2) (x2) 2.701(2) (x2) 2.627(2) (x2) 2.590(1) 2.622(2) 2.649(2) (x2) 2.633(2) (x2) 2.797(3) 2.562(1) 2.796(1) 2.604(2) (x2) Te(2)-Te(1 )-Cu(3) Te(2)-Te(1 )-Cu(4) Cu(3)-Te(1 )-Cu(3) Cu(3)-Te( 1 )-Cu(4) Cu(3)-Te(1 )-Cu(4) Cu(4)-Te(1 )-Cu(4) Te(1 )-Te(2)-Cu(1 ) Te( 1 )-Te(2)-Cu(2) Cu(1)-Te(2)-Cu(1) Cu( 1 )-Te(2)-Cu(2) Cu(1 )-Te(2)-Cu(2) Cu(2)-Te(2)-Cu(2) Te(3)-Te(3)-Cu(3) Te(3)-Te(3)-Cu(4) Cu(3)-Te(3)-Cu(3) Cu(3)-Te(3)-Cu(4) Cu(3)-Te(3)-Cu(4) Cu(4)-Te(3)-Cu(4) Te(4)-Te(4)-Cu(1 ) Te(4)-Te(4)-Cu(3) Cu(1 )-Te(4)-Cu(3) Te(6)-Te(5)-Cu(1 ) Cu(1)-Te(5)-Cu(1) 103.29(4) (x2) 104.65(4) (x2) 5901(7) 113.45(5) (x2) 152.06(5) (x2) 5907(7) 109.19(4) (x2) 106.02(4) (x2) 6269(7) 107.25(5) (x2) 144.73(5) (x2) 5996(7) 109.39(4) (x2) 107.16(4) (x2) 104.07(7) 6324(5) (x2) 143.43(5) (x2) 105.30(7) 103.60(3) 105.45(3) 1 1927(4) 104.66(4) (x2) 100.75(6) 1 Table 74. (cont'd) 221 Te(6)-Cu(2) Te(6)-Cu(2) Te(7)-Te(7) Te(7)-Cu(2) Te(7)-Cu(4) Te(8)-Te(8) Te(8)-Cu(1 ) Te(9)-Cu(3) Te(9)-Cu(4) Cu(1)-Cu(1) Cu(2)-Cu(2) Cu(3)-Cu(3) Cu(3)-Cu(4) Cu(4)-Cu(4) K(1)-Te(1) 2.631(2) (x2) 2.699(1) (X2) 2.797(3) 2.556(1 ) 2.594(1) 2.626(2) 2.626(2) (x2) 2.622(1) (x2) 2.625(1) (x2) 2.81 0(4) 2.625(3) 2.645(3) 2.769(2) 2.635(3) 3.703(3) (x2) Te(5)-Te(6)-Cu(2) Te(5)-Te(6)-Cu(2) Cu(2)-Te(6)-Cu(2) Cu(2)-Te(6)-Cu(2) Cu(2)-Te(6)-Cu(2) Cu(2)-Te(6)-Cu(2) Cu(2)-Te(6)-Cu(2) Te(7)-Te(7)-Cu(2) Te(7)-Te(7)-Cu(4) Cu(2)-Te(7)-Cu(4) Te(8)-Te(8)-Cu(1 ) Cu(1)-Te(8)-Cu(1) Cu(3)-Te(9)-Cu(3) Cu(3)-Te(9)-Cu(4) Cu(3)-Te(9)-Cu(4) Cu(4)-Te(9)-Cu(4) Te(2)-Cu(1 )-Te(4) Te(2)-Cu(1 )-Te(5) Te(2)-Cu(1 )-Te(8) Te(4)-Cu(1 )-Te(5) Te(4)-Cu(1 )-Te(8) Te(5)-Cu(1 )-Te(8) 106.36(4) (x2) 115.31(4) (x2) 105.60(6) 130.16(4) g“? 56.21 (7) 130.16(4) 6246(5) (x2) 105.66(3) 105.53(3) 109.60(4) 112.19(5) (x2) 6469(7) 6056(7) 9365(5) (x2) 6372(4) (x2) 6026(7) 104.86(4) 1 1054(5) 1 1365(6) 1 1592(6) 105.22(5) 106.56(5) Table 74. (cont'd) 222 K(1 )-Te(2) K( 1 )-Te(3) K(1 )-Te(3) K(1)- Te(4) K(1)- Te(5) K(1)- Te(6) K(1)- Te(7) K(2)- Te(4) K(2)- Te(5) K(2)- Te(5) K(2)- Te(7) K(2)-Te(8) K(3)-Te(2) K(3)-Te(5) K(3)-Te(6) K(3)-Te(6) K(3)-Te(7) K(3)-Te(8) K(3)-Te(8) K(4)-Te(3) K(4)-Te(4) K(4)-Te(9) K(4)-Te(9) 3.676(3) (x2) 3.715(3) 3.761(3) 3.719(3) (x2) 3.960(3) 3.796(3) 3.776(3) (x2) 2.546(3) (x2) 2.623(3) 2.996(4) 3.622(3) (x2) 3.613(3) (x2) 3.402(3) 3.756(3) (x2) 3.709(3) 3.976(3) (x2) 3.474(3) (x2) 3.347(3) 3.764(3) 3.677(2) 3.703(1) (x2) 3.636(3) (x2) 3.709(4) Te(2)-Cu(2)-Te(6) Te(2)-Cu(2)-Te(6) Te(2)-Cu(2)-Te(7) Te(6)-Cu(2)-Te(6) Te(6)-Cu(2)-Te(7) Te(6)-Cu(2)-Te(7) Te(1 )-Cu(3)-Te(3) Te(1 )-Cu(3)-Te(4) Te(1 )-Cu(3)-Te(9) Te(3)-Cu(3)-Te(4) Te(3)-Cu(3)-Te(9) Te(4)-Cu(3)-Te(9) Te(1 )-Cu(4)-Te(3) Te(1 )-Cu(4)-Te(7) Te(1)-Cu(4)-Te(9) Te(3)-Cu(4)-Te(7) Te(3)-Cu(4)-Te(9) Te(7)-Cu(4)-Te(9) 1 1066(6) 106.76(5) 1 1017(5) 9752(5) 1 1766(5) 1 11 .02(6) 103.76(5) 109.09(4) 1 1622(6) 1 1247(6) 109.45(5) 106.00(5) 105.69(5) 109.26(4) 1 1742(6) 1 1 1 .82(6) 109.65(5) 102.91 (5) 223 There are four crystallographically distinct K atoms in the unit cell of (l). The encapsulated K(1) atom is slightly off-centered on a rectangular plane defined by two opposite ditellurides in the dodecahedral cluster. The K(1)-Te distances range from 3.676 (3) A to 3.960(3) A (average 875(8) A). The remaining three K atoms are sitting in the large tunnels with various coordination environments {8 GM for K(2), 10 ON. for K(3), 6 ON. for K(4)}. The average K-Te distances are 370(16) A for K(2), 3.66(23) A for K(3), and 371(9) A for K(4). 3.2.2. Structure of Cs3Cu3Te1o (I) The structure of CS3003Te1o, an intriguing two-dimensional CuIT e framework, is shown in Figure 49. Interestingly, the two-dimensional CuIT e layer is composed of the fused dodecahedral Cua(Te2)6 clusters found in K40UgTe11. In this CuIl' e layer each dodecahedral cluster shares two sets of ditelluride edges out of three mutually perpendicular sets of ditellurides to form a layer as shown in Figure 50. The framework contains tetrahedral Cut centers bonded to Tezz' and Te2'. The Cu atoms in this layer are capped by quadruply bonded p4-Te2' ions above and below the layer. Based on the formula CsaCU3Te1o and on the assumption that the formal oxidation state on each ditelluride and telluride unit is -2, the formal oxidation state on Cu atoms is +1 (seven atoms) and +2 (one atom). ln CsacusTem all the copper atoms are crystallographically equivalent, implying complete delocalization of the one- electron vacancy (hole) on eight Cu atoms. This would result in the average formal oxidation state of copper being +1.125. However, in mixed-valence copper chalcogenide compounds, it is generally accepted that the formal oxidation state on the copper atom is +1. This is because coexistence of Cu2+ and Te2’ ions is thermodynamically unstable with respect to the electron 224 Figure 49. ORTEP representation of the unit cell of C33CU3T610. v/// ((11 Figure 50. ORTEP representation of (A) sideview and (B) top view of one [CuaTe1oln3n' layer. The Cua(Te2)5 and CugTea clusters are shaded for emphasis. 226 transfer from the reducing Q2' to the oxidizing Cu2+. If one then consider Cu atoms in +1 oxidation state, it follows that the hole resides on the tellurium atoms. Since Te-Te bond distances of T622' are in the normal range of single bond observed in the K4CuaTe11, the hole is expected to reside on the monotellurides. There are two crystallographically equivalent monotelluride atoms in the asymmetric unit. Making the simple assumption that the hole is delocalized on two monotelluride atoms, the resulting average formal oxidation state of monotelluride atoms is -1.5. This situation gives rise to a partially empty valence band consisting mainly of p orbitals of the two monotelluride atoms. As a result, this material should be a p-type metallic conductor. Based on this formal charge assignment, the structural formula can be better represented as Csa+Cua+(Te22')4Te21o5'. Certainly, these are only the formal oxidation states. The actual description is more complicated due to the small electronegativity difference between Cu (1.90) and Te (2.10) atoms. The geometry around the Cu atom is slightly distorted tetrahedral. The average Cu-Te distance is 263(5) A which is similar to that in (I). Short Cu-Cu contacts are also observed in this compound, ranging from 2.716(2) A to 2.744(2) A. The coordination environments of Te atoms are square pyramidal for Te(1)22', Te(2)22', and Te(4)2- and trigonal pyramidal for Te(3)22'. The average Te-Te distance of ditellurides is in the normal range of 281(2) A which is very close to that found in (I). The average Cu-Te bond distance of monotellurides (Te(4)) at 2605(1)A, which is slightly shorter than that found at 2.624(2)A in (I). Even though the difference is quite small, it is consistent with the reduced effective ionic radius of Te1-5'. Selected bond distances and angles are given in Table 75. 227 Table 75. Selected Bond Distances (A) and Angles (deg) in CsacuaTem with Standard Deviations in Parentheses Te(1 )-Cu Te(2)-Cu Te(3)-Cu Te(4)-Cu Te(1 )-Te(1 ) Te(2)-Te(2) Te(3)-Te(3) Cs(1 )-Te(1 ) Cs(1 )-Te(3) Cs(1 )-Te(3) Cs(1 )-Te(4) Cs(2)-Te(1 ) Cs(2)-Te(2) CS(2)-Te(3) Cu-Cu 2.670(1) (x4) 2.636(1) (x4) 2.594(1) (x2) 2.605(1) (x4) 2.606(2) 2.626(2) 2.765(2) 3.976(2) 4.160(1) (x4) 3.641(2) (x2) 3.604(1) (x2) 3.741(1) (x4) 3.799(1) (x4) 3.773(1) (x4) 2.744(2) Te(1 )-Cu-Te(2) Te(1 )-Cu-Te(3) Te(1 )-Cu-Te(4) Te(2)-Cu-Te(3) Te(2)-Cu-Te(4) Te(3)-Cu-Te(4) Te(1 )-Te(1 )-Cu Te(2)-Te(2)-Cu Te(3)-Te(3)-Cu Cu-Te(1 )-Cu Cu-Te(1 )-Cu Cu-Te(1 )-Cu Cu-Te(2)-Cu Cu-Te(2)-Cu Cu-Te(2)-Cu Cu-Te(3)-Cu Cu-Te(4)-Cu Cu-Te(4)-Cu Cu-Te(4)-Cu Cu-Cu 107.33 (5) 109.10 (5) 111.16 (5) 110.54 (5) 114.62 (5) 103.99 (5) 10746(3) (x4) 105.02(3) (x4) 107.06(3) (x2) 61.15(5) (x2) 107.61(6) (x2) 145.06(7) (x2) 6139(5) (x2) 110.15(5) (x2) 149.96(6) (x2) 106.65(6) 6356(6) (x2) 9566(7) (x2) 6265(6) (x2) 2.716(2) 228 There are two crystallographically distinct Cs atoms in the asymmetric unit. The encapsulated Cs(2) atom, situated on the crystallographic mmm site, is in the center of dodecahedral cluster and is surrounded by 12 Te atoms. The Cs(2)- Te distances range from 3.741(1)A to 3.779(1) A (average 877(3) A). The Cs(1) atoms are distributed between the [CsCuaTe1 012' layers. Within a radius of 4.0 A, Cs(1) has 5 Te atoms around it, whereas within a radius of 4.2 A it has 9 Te atoms around it. The Cs(1)-Te distances range from 3.804(2) A to 4.160(1) A, which is slightly longer than those of Cs(2)-Te. 3.3. Comparison of Cu3(Teg)e Cluster Size In K4CuaTe11 (I) and CsacuaTem (II) Two novel CuIT e compounds of K4CuaTe11 and CseCU3Te1o contain a common structural building unit, the dodecahedral CU3(Te2)5 cluster. At first glance, it is quite surprising that the Cs+ ion, whose ionic radius (2.02 A) is much larger than that of K+ ion (1.78 A), is encapsulated in the same dodecahedral cluster. Since the spatial requirements of the K+ and 06+ ions are expected to be very different, a change in volume of the cluster would be expected based on the volume of the cation encapsulated. However, a close comparison of the two Cua(Te2)6 dodecahedral clusters does not show appreciable differences in size and volume. Selected metric data for the two dodecahedral clusters are given in Table 76. 229 Table 76. Selected Metric Data for the KCua(Tez)6 Cluster (A) and the CsCua(Te2)6 Cluster (B) (A) (B) KCU3(Tez)6 CSCU8(T92)6 Dimensions(A) 6.822x6.955x7.091 6.936x7.013x7.053 (Based on Te22' edge) dav(Te-Te) (A) 2.606(11) 2.606(19) dmax(A-Te) (A) 3.960(3) 3.779(1) dmin(A-Te) (A) 3.676(3) 3.741(1) dav(A-Te) (A) 375(6) 377(3) dMCu-A) (A) 3.669(25) 3.735(1) dav(Cu-Te) (A) 263(4) 265(5) Volume (A3) of Cluster 145.9 150.8 of alkali ion 23.6 34.5 230 The average Te-Te, A-Te (A=K, Cs) and Cu-Te distances in both clusters do not show any appreciable differences. If we consider the interactions between alkali metal ions and tellurium atoms as ionic, the average K-Te distance of 3.75 A in (A) is very similar to the average Cs-Te distance of 877(3) A in (B). However, these K-Te and Cs-Te distances are not unusual and are in the normal range. These similar distances lead to almost negligible change in volume of the Cs-encapsulated dodecahedral cluster. Calculation of the entire volume of each cluster indicated a 4.9 A3 increase in (B). However, this volume change is still small compared to the increased volume (10.9 A3) of Cs+ ion compared to K+. If we assume that Cs+ ion fills completely the Cu/T e cluster then the K1” analogue should be slightly empty. This may be explained in terms of packing efficiency of the atoms in the cluster. Cluster (B), which contains three mutually perpendicular mirror planes, is more symmetric than cluster (A), which contains only one mirror plane, lying on atoms Te(3), Te(5), Te(6), and K(1). Therefore, cluster (B) is more close packed than cluster (A), giving the small volume change. Since K+ and 05+ ions are easily encapsulated inside the dodecahedral cluster, Rb+, whose ionic radius is in between them, should also be easily encapsulated. Indeed, RbacuaTeio containing an Rb-encapsulated dodecahedral cluster, was obtained recently in our Iaboratory"°. The 20-vertex dodecahedral Cua(Te2)6 cluster and 16-vertex CueTee cluster observed in this study are unique and, to the best of our knowledge, have no analogues in metal cluster chemistry. However, there is a 20-vertex dodecahedral cluster known in organic chemistry, the dodecahedrane, CononI. One can envision the possibility of building up novel extended structures in two or three dimensions based on this dodecahedral Cua(Te2)6 cluster by sharing more or all the ditelluride edges. The two novel structural 231 compounds, (I) and (II), containing Cue(Te2)6 clusters as building units highlighted the potential use of this dodecahedral cluster as a building unit for novel structural compounds. We expect to find more compounds based on this structural motif in the Gulf e system and further exploration on the AzTexICu systems is under way. 3.4. Charge transport properties of K4CuaTe11 (I) and C83CUaTe1o (II) Based on -2 formal oxidation states on ditelluride units and monotelluride atoms, each Cu atom has a formal oxidation state of +1. This gives rise to a completely filled valence band due to the d10 electronic configuration of Cu+, suggesting that K40U3Te11 should be a semiconductor. Preliminary conductivity measurements on several single crystals of K4CU3T611 show semiconducting behavior over the temperature range 30-300 K as shown in Figure 51. The conductivity increases from ~50 Slcm at 30K to ~160 Slcm at room temperature. However, the strong deviations from linearity of the log 0 vs. 1IT plot show a departure from classical semiconductor charge transport suggetsing that the charge transport is dominated by defects. The charge transport measurements on single crystals of C330u8Te1o along the (010) plane show that the resistivity at first decreases linearly with decreasing temperature and at low temperatures levels off to a constant value over the temperature range 5300 K, as shown in Figure 52. The resistivity increases from 1.2x10'6 Ocm at 5 K to 1.1x103 Ocm at room temperature. We note that the resistivity at 5 K is the smallest among the known ternary Cu chalcogenide compounds. It would be interesting to search for superconducting transition under pressure in this material. 232 $8.80.! .6 .993 can.» a .2 2293.5. .6 5.85. a mo San $.59 33:39.8 .8386 32.. Son. .5 2:90 :0: :82 on cm 3 N (ms/s) o 601 9N ... 233 6.8.8080 .o .325 case a .2 2220052 .o 3.8:... a am Sun €6.05 53.6.02 390 So". .mm 0590 C: 92900th :4... . _ ...... . . .....- 1.... _ . 4-91.. 1 1. com 1. 00¢ - com - com 1. coop comp (mo-I511) Al!/\!IS!SGH 234 The temperature dependence of the thermoelectric power (Seebeck coefficient) shows a very small positive value of 1~6 ,uV/K in the temperature range of 90~300 K as shown in Figure 53. The small and linearly increasing Seebeck coefficient with rising temperature indicates that CsacuaTem is a p-type metal as expected from the band filling argument advanced in the previous section. 235 98.030000 .o .9900 0.9.0 0 .o. 0.00 6.33 .0260 2.8209505 0.:.0.0aE0...0_nm_.0> .mm 0.390 O: 0.20.00E00. omm . 68 SN com cm. 2: on Ji «It... 11411.1...— l—ll 11111—11. — — _ u q d d 4 d A d l— - A d d P 111-11 .1111! ...iii...|l- i- ii , 1-! @ (x/Arl) Jemodouuaqi 236 3.5. Magnetic Susceptibility of Cs3Cu3Te1o (II) Variable temperature magnetic susceptibility data for CsaCuaTem over the temperature range 5-300 K are shown in Figure 54. When plotted as X" vs temperature, the data above 30 K show temperature independent paramagnetism, while below 30 K the data follow a Curie-Weiss Law. This behavior is characteristic of metals (Pauli paramagnetism) containing a low concentration of paramagnetic impurities. The temperature independent Pauli paramagnetism (xp=23.4 x10'5 emu/moi) has been obtained by subtracting the Curie-Weiss portion of the paramagnetism from the measured magnetic susceptibility. .._. 0> 350:9... 7.5. E00... .90....5080 05:020....58 .o. 0.00 208393 £33083 9.0.50... 0.20.0050.-030_.0> «6 0.39... 3.: 00m CON 2: o k a a a a a a a a a aaaaaflw N as an .../5 3.: ..V 2 0+ on 3 e. o a p b b n p p p F W a \I I9 W II m In 1.. I rm m a n \l a w a I W a O P (Ix IN V a m I I” . a . a .ee—x. a. . a .103 o— CHAPTER 7 Synthesis and Characterization of Mixed Chalcogenlde Compounds of K00432T0, K3CUaSqT02, and C017,5T03525 1. Introduction The synthesis and characterization of new materials is the essential step to understanding materials properties, discovering new properties, and often leads to the technological advances. We have been exploring new ternary metal chalcogenide compounds using alkali metal polychalcogenide fluxes as solvents and reagents at an intermediate temperature regime (150 < T < 500 0C). We have been quite successful in synthesizing novel structural compounds of (poly)sulfides, (poly)selenides, and (poly)tellurides of various transition metals in which a variety of new structural motifs have been discovered.5°-53-54 (see chapters 2-6) We believe that this molten salt synthesis method can be extended to other systems (e.g. a mixed-chalcogenide system) by combining, for example, K28 and Te or KgTe and S together. This approach could yield new materials with unusual structural features because of incorporation of two chalcogen atoms with different sizes atoms as well as coordination preference. In this chapter, we demonstrate our preliminary work in mixed chalcogenide systems. Two mixed-valence compounds, KCmSzTe and K30U384Te2 with interesting metallic properties, and one 238 239 three-dimensional compound, Cu17,6T63$25, were prepared in the mixed chalcogenide (K28/r e and KgTe/S) fluxes at 350-450 0C. 2. Experimental Section 2.1 Reagents Chemicals were used as obtained: copper powder, electrolytic dust, purified, Fisher Scientific 00., Fair Lawn, NJ; tellurium powder, -100 mesh, 99.95% purity, Aldrich Chemical 00., Milwaukee, WI; sulfur powder, sublimed, J. T. Baker Chemical Co., Phillipsburg, NJ; potassium metal, analytical reagent, Mallinckrodt Inc., Paris, KY 2.2. Physical Measurements Magnetic susceptibility measurements were performed on a MPMS Quantum Design SQUID magnetometer over the temperature range 5 K to 300 K for KCU482Te (at 5 kG) and K30u884Te2 (at 7.5 kG) and over the temperature range 2 K to 300 K for Cur/,6Te8826 (at 100 G). Single crystals of each compound were manually selected for measurements and were used without grinding as random oriented single crystals. The data for KCU482Te and K30U384Tez were corrected for diamagnetic contributions of the sample holder. To obtain molar susceptibility the corrections for ion-core diamagnetic contributions from atomic constituents of KCU4SQTG and K3Cuas4Te2 were made using the values tabulated by Mulay103. The magnetization of each compound was examined as a function of applied field from 250 G to 8 kG for KCU482Te and K3Cuas4Tez (at 5 K) and from 200 G to 9.5 kG for 240 Cu17,6Te8823 (at 2 K). The magnetization of KCU432T6 and K3CueS4Te2 was found to vary linearly with the applied field, while the magnetization data for Cu17,6Te3$26 showed satutration behavior above 3 kG. It started to converge at 600 G and remain constant above 3 kG as shown in Figure 55. 241 .... 6.0.. 9.9.9.... 8.68 .o 5.5:... a mo 8008.250 0c__.0.0>.o>_oa .9 0.00 3E0. 3:030:00... .mm 0.30.0 .0..= . . . . . (MO -N W a B 5 U I 1V W... O a U a a some m I I I a I 10 l\ I I I 4.25 m 242 Four-probe dc resistivity and thermoelectric power data for KCU4SzTe were provided by Prof. Carl R. Kannewurf (Northwestern University) over the temperature range 5 K to 300 K. A computer automated measurement system was employed to obtain thermopower and resistivity data with both the current and thermal gradient applied along the (010) plane for KCU482T6. For all measurements, electrode connections to the small single crystals were made with the use of 25 and 60 pm gold wires and gold bonding paste. Quantitative microprobe analyses of the compounds were performed on a Jeol 35CF scanning electron microscopy equipped with Tracor Northern TN 5500 X—ray microanalysis attachment. Single crystals of each sample were carefully picked and mounted on an aluminum stub using conducting silver paint to help dissipate charges that developed on the sample surface during measurements. Energy Dispersive Spectra (EDS) were obtained using the following experimental set-up: X-ray detector position : 55 mm Working distance :39 mm Accelerating voltage : 20 KV Take-off angle : 27 deg Beam current : 200 picoamps Accumulation time : 100 seconds Window : Be A standardless quantitative analysis (SQ) program was used to analyze the X- ray spectra obtained. Since the Cu ratio is always overestimated due to the contribution from system Cu peaks, a correction factor (x 0.73), determined by calibrating known K/Cufl’ e compounds, was used to evaluate Cu percentage. 243 2.3. Synthesis Chemicals were measured and loaded in Pyrex tubes under a dry nitrogen atmosphere in a Vacuum Atmospheres Dri-Lab glovebox. Potassium monosulfide (K28) and potassium monotelluride (K2Te) were prepared in liquid ammonia from potassium and elemental sulfur (or tellurium) in a 2:1 ratio. Potassium blsflu-sulfldo)(ua-telluro)tetracuprate(l,ll), KCuqszTe (I) 0.165 g (1.5 mmol) of K28, 0.048 g (0.75 mmol) of Cu and 0.510 g (4.0 mmol) of Te were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 450 °C for 4 days and cooled slowly to 100 °C at a rate of 2 °C/hr then to 50 °C at a rate of 10 oClhr. Black rectangular-shaped crystals were obtained by removing excess potassium polychalcogenides (K2Tex, Kzsy, KzTenSm) with DMF under a N2 atmosphere (yield: 83.6 % based on Cu used). A quantitative microprobe analysis performed on a large number of crystals with the EDS/SEM system gave an average composition of KCU3,782,oTe1_1. Tripotassium tetra(p4-sulfldo)bis(p5-telluro)octacuprate(l,ll), Kacuasfleg (ll) 0.165 g (1.5 mmol) of K28, 0.032 g (0.5 mmol) of Cu and 0.319 g (2.5 mmol) of Te were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10'3 torr). The tube was placed in a computer-controlled furnace and heated at 450 0C for 4 days and slowly cooled to 100 °C at a rate of 2 °CIhr. then to 50 °C at a rate of 10 °CIhr. Black needle- like crystals were obtained by removing excess potassium polychalcogenides (K2Tex, K28y) with DMF under a N2 atmosphere (yield: 71.4 % based on Cu used). A quantitative microprobe analysis performed on a large number of 244 single crystals with the EDS/SEM system gave an average composition of K30U8.3S4.2Te1.9- Copper tellurlum sulflde, Cu17,31'eas25 (lll) 0.103 g (0.5 mmol) of K2Te and 0.032 g (0.5 mmol) of Cu and 0.128 g (4.0 mmol) of S were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10:3 torr). The tube was placed in a computer-controlled furnace and heated at 350 °C for 4 days and cooled slowly to 100 °C at a rate of 2 °CIhr. then to 50 °C at a rate of 10 °CIhr. Black cubic-shaped crystals were obtained by removing excess potassium polychalcogenides with DMF under a N2 atmosphere. The products were contaminated with some amounts of yellow needle crystals of K2Te83107; these were easily removed by washing with water (yield: 45 °/o based on Cu used). A quantitative microprobe analysis performed on a large number of single crystals with the EDS/SEM system gave an average composition of Cu17Te7,SS26. 2.4. x-ray Crystallographic Studles All compounds were examined by X-ray powder diffraction for phase characterization and identification. The d-spacings for each compound were obtained from powder pattern recorded on a Phillips XRG-3000 computer- controlled powder diffractometer, operating at 40KV, 35 mA. Graphite monochromated Cu radiation was used. To verify product homogeneity, the d- spacings observed for the bulk materials were compared, and found to be in accord, with those calculated from the single crystal X-ray structure analysis data. The calculation of d-spacings was performed using the POWD10 program56. The results are summarized in Tables 77-79. 245 Table 77. Calculated and Observed X-ray Powder Diffraction Pattern of KCU482Te H K L awed) amok) lllmax(obs.) 0 0 1 10.22 10.16 94.9 0 0 2 5.11 5.09 23.0 1 0 o 3.67 3.67 9.3 1 0 1 3.62 3.63 16.3 0 o 3 3.40 3.40 21.5 1 0 2 3.06 3.06 69.7 1 0 3(0 0 4) 2.55 2.55 40.2 1 1 2 2.415 2.406 51.3 1 1 3 2.135 2.131 100 0 0 5 2.044 2.042 49.1 2 0 0 1.9361 1.9372 37.0 1 1 4 1.6669 1.6636 16.6 1 0 5 1.6060 1.6056 20.1 0 o 6 1.7033 1.7023 65.4 2 1 2 1.6416 1.6392 26.0 246 Table 78. Calculated and Observed X-ray Powder Diffraction Pattern of K3CU384T62 H K L awed.) comm lllmax(obs.) 0 0 1 9.66 9.63 15.2 -2 0 1 7.33 7.25 100 0 0 3 3.22 3.23 22.6 -4 0 3(6 0 0) 2.91 2.91 46.6 5 1 0 2.60 2.60 29.6 1 1 3 2.409 2.403 33.7 -5 1 3 2.212 2.211 15.2 -7 1 1 2.142 2.140 34.6 -1 1 4 2.090 2.066 31.2 6 0 3 (0 2 0) 1.962 1.962 34.6 -7 1 4(-9 1 2) 1.7572 1.7500 22.1 -8 0 5 1.6363 1.6390 31.2 247 Table 79. Calculated and Observed X-ray Powder Diffraction Pattern of CU17.6T86$26 H K L awed) comm) l/lmax(obs.) 2 0 0 5.11 5.11 11.1 2 2 0 3.61 3.62 31.4 2 2 2 2.95 2.96 100 3 2 1 2.73 2.74 16.6 4 0 o 2.55 2.56 20.6 3 3 0(4 1 1) 2.413 2.411 12.5 4 2 o 2.269 2.229 14.7 4 2 2 2.069 2.092 16.1 5 1 0(4 3 1) 2.007 2.009 7.9 5 2 1 1.6692 1.6710 10.9 4 4 0 1.6099 1.6129 71.5 4 3 3 1.7556 1.7567 6.6 4 4 2 1.7064 1.7079 9.5 6 1 1 1.6609 1.6635 12.5 6 2 0 1.6166 1.6209 12.0 6 2 2 1.5435 1.5456 44.3 J . air—TY 5} 1' 248 The single crystal data of KCU4S2Te were collected on a Nicolet P3 four circle diffractometer with graphite monochromated Mo-Ka radiation using the 0-20 scan mode. The single crystal data of K30U384Te2 and Cu17,eTe3826 were collected on a Rigaku AFCGS diffractometer with graphite monochromated Mo-Ka radiation using the (1)-20 scan mode. Accurate unit cell parameters for all compounds were obtained from least-squares refinement of the 20, m, X. and 4) values of 20-25 machine-centered reflections. The stability of the experimental setup and crystal integrity were monitored by measuring three standard reflections periodically (every 100 reflections) during data collection. The intensities did not show any appreciable decay. Two absorption corrections were applied to the data of all compounds: an empirical absorption correction based on 1): scans for three reflections followed by a DIFABS57 correction. The structure of (l) was solved with direct methods using SHELXS-8658 and was refined with the SDP59 package of crystallographic programs. The structures of (II) and (Ill) were solved with direct methods using SHELXS-86 and were refined with the TEXSAN60 package of crystallographic programs. All calculations were performed on a VAXstation 3100 computer. During the isotropic structural refinement of (III), high temperature factors of 9.678 and 3.881 A2 are observed for Cu(2) and 8(2) atoms respectively implying a partial site occupancy of these atoms. Thus, the site occupancies of Cu(2) and 8(2) were refined individually by fixing every other atom position. The refined Cu(2) site occupancy was 46.64 % of that allowed, whereas the refined 8(2) site occupancy was 105.53 %. Since this 8(2) atom site occupancy is physically unrealistic, it was fixed at 100 %. Upon fixing the site occupancy of Cu(2) atom at 46.64 %, further isotropic refinement followed by DlFABS absorption correction gave an RlFlw value of 4.1146 with Beq of Cu(2) and 8(2) being 2.25 A2 and 4.61 A2, respectively. Even though the temperature 249 factor of S(2) atom is large, anisotropic refinement of all atoms was carried out, resulting in final R/Rw value of 2913.8. If the 8(2) site occupancy is adjusted downwards by creating a small vacancy, we obtain a higher minimum FllFfw value of 3.0/4.1. Therefore, full site occupancy of S(2) seem reasonable even though its temperature factor is high at 4.587 A2. Based on this refinement, nonstoichiometric structural formula of Cu17,6Te3S26 was obtained. The complete data collection parameters and details of the structure solution and refinement for (I), (If), and (III) are given in Table 80. The final atomic coordinates, temperature factors and their estimated standard deviations are shown in Tables 81 -83. 7] 250 Table 80. Summary of Crystallographic Data for KCU482Te, K30U8S4T62, and CU17.6T96$26 comm I II III Formula KCU432Te K3CU3$4T82 CU17_6T63$26 Formula weight 484.99 1009.10 2966.13 space group P4Immm C2Im l-43m a(A) 3.6762(7) 17.936(3) 10.2385(6) b(A) 3.8762(7) 3.912(3) 10.2385(6) c(A) 10.220(3) 9.924(2) 10.2385(6) 0: (deg) 90.0 90.0 90.0 8 (deg) 90.0 102.58(2) 90.0 7 (deg) 90.0 90.0 90.0 Vol (A3), 2 153.6(1), 1 679.6(8), 2 1073.270), 1 Temperature (°C) 23 23 23 Crystal size (mm) 0.13x0.13x0.05 0.57x0.05x0.05 0.13x0.10x0.08 Radiation Mo-Ka Mo-Ka Mo-Ka u (Mo-Kc, cm‘1) 195.9 180.2 152.8 Dcalc (glcm3) 5.24 4.93 4.59 28".” (deg) 50 50 55 Scan method 8/28 (1)/20 (1)/20 No. of data collected 627 1241 510 No. of unique data 117 697 258 No. of data used 107 589 222 (F02>30(F02)) No. of atoms 4 9 5 No. of variables 13 53 1 8 Phasing technique Direct methods Direct methods Direct methods Final RIFtW (%) 2.96/3.15 3.7/56 2913.8 Max. shift/esd 0.0 0.0 0.0 (last cycle) Extinction coefficient 1.93x10°6 N/A 6.09x10'5 ¥ :lf-Hl .' ”MKS 251 Table 81. Fractional Atomic Coordinates and Beq Values for KCu4S2Te with Their Estimated Standard Deviations in Parentheses Atom x y z Beq“. A2 Te 0 0 0 093(2) Cu 1/2 0 0.8101(2) 1.67(3) K 0 0 1/2 1 .74(8) S 1/2 1/2 0.6882(5) 1 . 10(6) a B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Baq = (4I3)[aZB11 + sz22 + 02833 + ab(cos 10812 + ac(cos 0813 + bcloos 00823]. Table 82. Fractional Atomic Coordinates and Values for lQCu884Te2 with Their Estimated Standard Deviations in Parentheses Atom x y 2 Beqa. A2 Te 0.33526(5) 0 0.8650(1) 1 .44(4) Cu(1) 0.481 1 (1 ) 0 0.8508(2) 1.42(7) Cu(2) 0.3882(1) -1/2 0.7150(2) 210(8) Cu(3) 0.2960(1) 0 0.5642(2) 224(9) Cu(4) 0.4217(1) -1/2 0.9838(2) 1 .69(8) S(1) 0.1569(2) 0 0.5158(3) 0.8(1) S(2) 0.5293(2) -1 l2 0.7860(3) 0.8(1 ) K(1) 1/2 0 1/2 1 .7(2) K(2) 0.3274(2) 0 1 2222(3) 1 .7(1) 8‘ 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[aZB11 + b2822 + 62633 + ab(cos 0812 + ac(cos (3)313 + belcos «98231 252 Table 83. Fractional Atomic Coordinates and Beq Values for Cu17,sTe3826 with Their Estimated Standard Deviations in Parentheses Atom x y z Baqa, A2 Te 0.25978(8) 0.2598 0. 2598 1 .1 81 2(4) Cu(1) 1/4 1/2 0 2.4(1) Cu(2) 0.2149(7) 0 0 2. 7(2) S(1) 0.1106(3) 0.1106 0.3576(3) 1.60(9) S(2) 0 0 0 4.567(5) a 8 values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as Beq = (4/3)[aZB11 + b2322 + 02333 + ab(cos 0812 + ac(cos (3)813 + ac(cos 08231. 253 3. Results and Discussion 3.1. Synthesis Synthesis of the mixed chalcogenide compounds of KCU4S2Te and K30U3S4Te2 has been readily achieved using mixed-chalcogenide fluxes (K2S/l'e) as solvents and reagents at 450 0C as shown in eq 1. nK2S + mCu + qTe ------ > KCU4S2Te or K3CU3S4Te2 eq 1. where the ratios of n/m/q are 3/1.5l8 and 3/1/5 for (l) and (II) respectively. Surprisingly, when we used another possible combination of mixed- chalcogenide flux (K2Te/S) at 350 0C as shown in eq 2, K2Te + CU + 83 ------> CU17,6T83826 eq 2. the new structural compound of Cu17,6Te3$26 was obtained. The equations given above are not balanced and only show the reactants used (in left-hand side) and products isolated (in right-hand side). After reaction, excess mixed- chalcogenide fluxes were easily removed from the products by washing with DMF. All compounds are stable in the air and moisture for several days. During the study of new ternary chalcogenide materials in the A2Ox flux, it came to our attention that it is possible to have mixed-chalcogenide fluxes by simply combining, for example, K28 and Te or K2Te and S. We expected new structural compounds in this mixed chalcogenide system due to the incorporation of two chalcogenides with different size into the lattice. We chose ‘fl—- 1", 254 S and Te because the combination of S and Se or Se and Te, which are similar in size, might result in poSitional disorder between them. In the beginning small crystals of KCU4S2Te were obtained in the reactant ratios K2SICu/T e of 1~4l118 at 350 0C with contaminations of elemental tellurium and a few yellow needle-like crystals of known K2Te83112. Since the K2T683 phase was soluble in water, it was easily removed from the product by washing with water. We varied reaction conditions in order to increase product homogeneity and single crystal size for charge transport property measurements. Upon changing Cu metal ratios and the reaction temperature, large single crystals (up to 2mm) of KCU482Te were obtained as a pure phase from the 3115/8 ratio at 450 DC. A parallel approach used in an effort to obtain a homogeneous KCU4S2Te phase, was to lower the Te ratio at 450 00. We thought this might suppress the elemental Te in the final product. However, it resulted in unexpected isolation of new mixed chalcogenide compound K3CuaS4Te2 in high yield from the K2S/Cu/T e ratios of 3/1/5~6. In addition to the two new mixed chalcogenide compounds (I) and (If) there is at least one more promising phase in this system. Golden brown needle-like crystals were obtained as a minor phase from the 3/1/8 ratio at 350 00 along with KCU4S2Te. These crystals have the approximate compositions of KCU2TeS from the EDSISEM analysis. Unfortunately, attempts, to grow large single crystals of this new phase suitable for X-ray single crystal study, were unsuccessful. This new phase seems to compete with KCU4S2Te at 350 0C, but is completely suppressed by increasing the reaction temperature to 450 oC. Another choice for the mixed-polychalcogenide flux (K2TelS) showed redox chemistry associated with Tez"4+. In the sulfur-rich K2Te/S flux, sulfur atoms oxidize most of Te2' to Te(1+ while they are reduced to S2: As a result, [T eS312' species are formed in the flux as shown in eq 3. —___ 3.1M ' 255 K2Te + 88 -----> K2[T683] + K2Sx eq3. The [TeS312' species in the flux can act as ligands to Cu+ atoms to form Cu17,5TeaS26. [TeS312' ions are also crystallize with K+ ions as K2TeS3. Fortunately, they are quite soluble in water and can be easily removed by washing with water. [T833]2' species are not formed in the K28/T e flux because S2' is already completely reduced species and can not oxidize Te atoms. 3.2. Description of Structures 3.2.1. Structure of KCu482Te (I) KCU4S2Te is isostructural to the known mixed-valence compound KCU48361. The structure of KCU4S2Te is shown in Figure 56. A tellurium atom replaces the eight-coordinate sulfur atom in KCU4S361la). The structure is composed of double layers and charge compensating K+ ions. The [Cu4S2Teln ' double layer is made of fused anti PbO-type CU2STe layers. The plane composed of tellurium atoms along the (010) plane coincides with a mirror plane which bisects the [CU4S2Te]n"' double layer. This double layer is shown in Figure 57. Based on the formula KCU4S2Te and on the assumption that the oxidation state on each S and Te atom is -2, the formal oxidation state on copper is +1 (three atoms) and +2 (one atom). However, the coexistence of Cu2+ and 02' is thermodynamically unstable with respect to electron transfer from the reducing 02' to the oxidizing Cu2+. Thus, all copper atoms are +1 and the one electron vacancy (hole) resides on chalcogen atoms. The known mixed-valence compound KCU4S3 was also formulated as 256 0' '3. U) 3:5 OS X ._m CE :42 ‘55 3.0 00 E'— (30 A .2; E 3.9 as: as 322 93 52 AA 9:5 t: 5.5 A =8 8". =0. g9 09 58 "> $0 3E c: a) .90: En =8 8.1: a): no a. A 103. < “E v m: ”‘0 "c (1:... 0‘6 'C 82§ L— 90 :50) .9: [LI— 257 Figure 57. ORTEP representation of the [CU4S2Telnn' layer. Black circles are Te, open circles are S, and crossed circles are Cu atoms. 258 K+(Cu+)4(82')2S'.61(a) Since there is only one crystallographically distinct Te atom in the asymmetric unit and this is more likely to be oxidized compared to S atom, the hole probably resides on this Te atom. This results in -1 formal oxidation state of Te, creating a partially empty valence band. Thus the chemical formula can be represented as K+(Cu+)4(S2')2Te'. Of course, these are only the formal oxidation state. The actual description is more complicated due to small electronegativity difference among Cu, S, and Te atoms. The Cu atoms are situated on a mm crystallographic site and have distorted tetrahedral geometry. They are bonded to two sulfur atoms and two tellurium atoms. The sulfur atoms, which are situated on a 4 m m crystallographic site, are bonded to four Cu atoms and have a square pyramidal geometry. The Cu-S-Cu bond angles of 73.1(1)o and 114.5(2)° are comparable to those found in KCU4S361la). The tellurium atom on the 4/mmm site is bonded to eight Cu atoms with a square prismatic geometry. The Cu-S bond distance of 2.304(3) A is very close to that (2.312(2)A) of KCU483. If one considers the effective ionic radius of Tel: to be smaller than that of Tez', the Cu-Te bond distance should be shorter than the normal Cu-Te bond distance. However, the Cu-Te bond distance is in the normal range of 2.743(1) A. This is probably due to the higher coordination number (8) of Te atoms. The average S-Cu-S bond angle is 114.5(2)°, comparable to that found in KCU483 (115.0(1)°). However, the Te-Cu-Te bond angle of 89.92(5)° strongly deviates from the ideal tetrahedral angle. This is because substitution of a Te atom for a S atom in the parent compound KCU483 did not cause significant changes in the cell parameters along the a- and b-axis. One would have expected tellurium substitution to increase in cell parameters of (l) due to its larger size. However, similar chemical environments and spatial requirements of K+ ions in both compounds prevent the cell parameters from changing along the a- and b- 259 axis. Thus, to accommodate a longer Cu—Te bond, the cell increases only along the crystallographic c-axis, resulting in a smaller Te-Cu-Te bond angle. There are short Cu---Cu contacts at 2.741(1) A, very similar to that (2.757 A) of KCU483. The Te-Te contacts of 3.8762(7) A along the a- and b-axis are too large to be considered as a significant Te-Te interactions . Selected bond distances and angles are given in Table 84. There is only one crystallographically distinct K+ ion situated on a 4Immm site. It is surrounded by eight sulfur atoms in a square prismatic geometry. lt participates in ionic interactions only with S atoms. The average K--S distance is 3.349(3) A which is close to that (3.351(2) A) of KCU48361(a). Hm- ‘ L7" _. 260 Table 84. Selected Bond Distances (A) and Angles (deg) in KCU4S2Te with Standard Deviations in Parentheses Te-Cu 2.743(1) (x6) Cu-Te-Cu 6992(4) (x4) Cu-Te-Cu 9009(4) (x4) Cu-Te-Cu 180.(0) (x4) Cu-Te-Cu 5995(2) (x8) Cu-Te-Cu 120.05(2) (x8) Cu-S 2.304(3) (x2) Cu-S-Cu 1 14.5(2) (x2) Cu-S-Cu 73.1(1) (x4) Cu-Cu 2.741 (1 ) Te-Cu-Te 8992(5) Te-Cu-S 1 1249(6) (x4) K-S 3.349(3) (x8) S-Cu-S 114.5(2) l 261 3.2.2. Structure of K30u384Te2 (II) K3CU3$4T82 is isostructural to the known mixed valence compound K3Cuass71. The structure of K3Cu8S4Te2 is shown in Figure 58. It is composed of two-dimensional layers and charge balancing K+ ions. The layer of [Cu884Te2Jn3n' is made of two easily recognizable fragments: (a) [CU4Q41n3n' type column found in NmCU4Se4 (see Figure 11) and (b) the anti PbO-type Cu() layer found in NaCuTe (see Figure 41). These two fragments are fused together to form the [CU3S4Te2ln3n' layer. The [Cues4Te2ln3n' layer is shown in Figure 59. Based on the formula K3CLleS4T62 and on the assumption that the oxidation state on each S and Te atom is -2, the formal oxidation state on copper would be +1 (seven atoms) and +2 (one atom). However, Cu2+ ions are easily reduced to Cu+1 in the presence of reducing 02‘ ions. Thus, all copper atoms are +1 and the one electron vacancy (hole) resides on the chalcogen atoms. The known mixed-valence compound K3CU3$5 was formulated as K+3(Cu+)3(82‘)5S1'.71 Since there are two crystallographically equivalent Te atoms in the asymmetric unit and the Te atom is more likely to be oxidized compared to the S atom, the hole is expected to delocalize on the Te atoms. This gives -1.5 formal oxidation state for each Te atom, creating a partially empty valence band. Thus, the structural formula can be represented as K+3(CU+)8(SZ')4(T91'5‘)2- There are four crystallographically distinct Cu atoms in IQCuaS4Te2. Two Cu atoms, Cu(1) and Cu(4), in the [CU4O4ln3n' column have distorted trigonal planar geometry. Cu(1) atoms are bonded to two S(2) atoms at 2.287(2) A and one Te atom at 2.650(2) A. Cu(4) atoms are bonded to one S(2) atom at 2.262(4) A and two Te atoms at 2.612 (2) A. These Cu-Te bond distances are comparable to that of the three coordinate Cu atoms in -.:.SI 1 262 Figure 58. ORTEP representation of the unit cell of K30u3$4Te2. 263 Figure 59. ORTEP representation and labeling scheme of the [CuaS4Te21n3fl' layer. 264 NaCuaTe297 (2.575(8) A). Two other Cu atoms (Cu(2) and Cu(3)) in the CuQ layer fragment have distorted tetrahedral geometry. Cu(2) atoms are bonded to two S atoms and two Te atom while Cu(3) atoms are bonded to three S(1) atoms and one Te atom. The average Cu—S bond distances are 237(15) A for Cu(2) and 237(6) A for Cu(3) which are very close to those of K3Cuase71 (2.34(15) A and 232(4) A, respectively). The Cu-Te distances are 2.749(3) A for Cu(2) and 2.914(3) A for Cu(3). This Cu(3)-Te bond distance of 2.914(3) A is slightly longer than the normal Cu-Te bond distances of known compounds with four coordinate Cu atoms. In KgCuase, the Cu-S bond distance corresponding to this Cu(3)-Te bond is much longer at 2.84 A compared to the normal Cu-S bond distance of 2.30 A. There are short Cu-Cu contacts in the range of 2.604(3)~2.779(2) A. There are two crystallographically distinct sulfur atoms and one tellurium atom in the asymmetric unit. Both sulfur atoms have square pyramidal geometry while the tellurium atom has 6 Cu atoms around it as shown in the following scheme. 265 This tie-coordination of the chalcogen atom is unusual even though it is found in similar structural compound of KCU3S270. The average Cu-Te distance is 2.71(12) A. Selected bond distances and angles are given in Table 85. There are two crystallographically distinct K atoms in the asymmetric unit located between the [CuaS4Te2]n3"' layers. The K(1) atom has 4 sulfur atoms around it. The shortest K-S bond distance is 3.392(3) A. This K(1) atom does not have any interaction with Te atoms. The K(2) atom has 3 Te atoms and 4 S atoms around it. The shortest K-Te and KS distances are 3.412(3) A and 3.127(4) A respectively. 266 Table 85. Selected Bond Distances (A) and Angles (deg) in KacuaS4Te2 with Standard Deviations in Parentheses Cu(1)-S(2) Cu(2)-S(1) Cu(2)-S(2) Cu(3)-S(1) Cu(3)-S(1) Cu(4)-S(2) Cu(1 )-Te Cu(2)-Te Cu(3)-Te Cu(4)-Te Cu(1)-Cu(2) Cu(1)-Cu(4) Cu(2)-Cu(3) Cu(2)-Cu(4) Cu(3)-Cu(3) Cu(4)-Cu(4) 2.267(2) (x2) 2.257(4) 2.476(4) 2.435(4) 2.337(3) (x2) 2.262(4) 2.650(2) 2.749(2) (x2) 2.914(3) 2.612(2) (x2) 2.726(2) 2.705(2) 2.779(2) 2.604(3) 2.693(3) 2.757(4) Cu(2)-S(1 )-Cu(3) Cu(2)-S(1)-Cu(3) Cu(3)-S(1)-Cu(3) Cu(3)-S(1)-Cu(3) Cu(1)-S(2)-Cu(1) Cu(1 )-S(2)-Cu(2) Cu(1 )-S(2)-Cu(4) Cu(2)-S(2)-Cu(4) Cu(1 )-Te-Cu(2) Cu(1 )—Te-Cu(3) Cu(1 )-Te-Cu(4) Cu(4)-Te-Cu(4) Cu(2)-Te-Cu(2) Cu(2)-Te-Cu(3) Cu(2)-Te-Cu(4) Cu(2)-Te-Cu(4) Cu(3)-Te-Cu(4) Te-Cu(1 )-S(2) S(2)-Cu(1)-S(2) Te-Cu(2)-S(1 ) Te-Cu(2)-S(2) Te-Cu(2)-Te S(1)—Cu(2)-S(2) 109.0(2) 74.4(1) (x2) 113.6(2) .-...- 66.7(1) (x2) 1 17.6(2) 6977(9) (x2) 76.6(1) (x2) L 1 15.8(2) 6066(5) (x2) 66.1 1(7) 6166(5) (x2) 9696(8) 90.71 (8) 5669(5) (x2) 5606(6) (x2) 122.430) (x2) 116.75(5) (x2) 116.93(9) (x2) 1 17.6(2) 117.38(8) (x2) 107.35(7) (x2) 90.71 (8) 1 14.0(2) II .11" “i- Table 85. (cont'd) 267 K(1)-S(1) K(1)-S(2) K(2)-S(1 ) K(2)-S(2) K(2)-Te K(2)-Te K(2)-Te 3.402(3) (x2) 3.392(3) (X2) 3.217(4) (x2) 3.245(4) (x2) 3.579(4) 3.462(3) 3.462(3) Te-Cu(3)-S(1 ) Te-Cu(3)-S( 1 ) S(1)-CU(3)-S(1 ) S(1)-Cu(3)-S(1) Te-Cu(4)-Te Te-Cu(4)-S(2) 102.2(1) 106.9(1) (x2) 111.3(1) (x2) 1 13.6(2) 5”“: 9696(6) ' 121.19(7) (x2) 268 3.2.3. Structure of Cu17,5TeaS25 (Ill) Structually, the compound Cu17,6T63S26 is related to the tetrahedrite mineral class and can be represented by the general formula M12X4S13 (M=Cu, Ag, Zn, Cd, Fe, Hg; X=Sb, As, 80‘”. Cu17,6T63S26 has a group lVB metal (Te atom) occupying the X site in M12X4S13. The structure of Cu17,5TeBS26 can be viewed as the derivative of the sphalerite structure (ZnS). One complete unit cell (Cu12Te3824)(Cu68) is shown in Figure 60. The Te is formally in the +4 oxidation state. In this structure there are easily recognizable TeS32' units. These TeSaz' units, which are isoelectronic to 8032', bridge Cu atoms via sulfur atoms to form a three dimensional framework similar to that in sphalerite, but with large voids inside it. In fact, Te atoms occupy the tetrahedral Zn site in sphalerite but are bonded to only three sulfur atoms. The fourth coordination site is occupied by the Te lone electron pair. This results in relatively large empty cages (cavities) in the structure. These cavities are then filled with Cues units as in the tetrahedrite structure. In tetrahedrite there are two CusS units on the (0, 0, 0) and (1/2, 1/2, 1/2) positions per unit cell. However, the substitution of Te4+ for Sb3+ creates an electron deficiency in the framework resulting in the 53.3 % vacancy of Cu(2) atoms in Cues. Therefore, the voids are actually filled with CU2,3S units instead of CusS units, with the 2.8 Cu atoms occupying equally the 6 possible sites. Thus, the structural formula unit of Cu17,5Te3826 can be reformulated as [Cu12(TeS;3)3](CU2,3S)2. Based on this nonstoichiometric structural formula and on the assumption that the formal charge of Te and S are +4 and -2 respectively, a nonintegral oxidation state is expected for Cu. There are two crystallographically distinct Cu atoms in the asymmetric unit. The Cu(1) atom in the [Cu12(Tesa)8] cage has slightly distorted tetrahedral geometry with four S(1) atoms around it. The average "I'i )1 ~ A ‘.' f 911166461161 art! of nonfatal“. “Bait grangtM slumtot letsnejg 971! yd m éi'v‘l 0001988811 39886151in .8"(ta .BA .‘fix ,I“ I‘” 1‘ 912-431.1113 artT 5.1893.er nt 6112 X ertt m ( r 71'." ' .11». 53:19 .2113: “m1 f". avlfs‘fltilb 3:1! as W N f.) , .rw ' 1,3,072‘l;_«;?'3eeTt_-pua,lIbotlrtu -. ,1: nuts, 9:, 211‘ hi this noifsbixohflIQ 1: :, It; :“1; T11117\"":1l T989711- MW 1 . 1,“. i, ”5141;: m3.) ofemOIIWIN“ :1: 11f, '14! ;‘ 1.5: ififlSthUd.“ ¢_ .' Linux ..‘m "(3 3918.341” "I * ¢ 1 7 u, 1,; 7. 7'7: , 0»: .3310 s: airs M r,” ‘ ytqme OW m 21,2 _.1 , -, ,5 , 1.191.1- ‘l er. almu 8900 M.” ms -' ‘ r «ii "‘~'-;) 31inu 8.00 N I '7r.l.°’.l1ledU2 9111.1.“ ' .. . ~ ,_ ,, 1* 01001293331710“ >- _ .3 -. .. 93,. , .. ._ 4. 3,3,1: , . 11:1. l‘“l.§l\.\(1ISU?OS sum-I “3'13 . ‘fiz, '1 -.¢ 7 ,- ,'..~ '5‘ #22" L E3 .—;i"?' V’lbl'C’i‘l QD‘YQUOOO m fl rig/FE: ‘ ,,3‘;, 1031‘“ .336, f 1.7;); 2),, ,9., mats" 5C? 3333 31335873er U“ tgril It; " 1.135323 9:11 at), (we slurrug'r ';:i:il.i'.i12€ enfsmottloiofenon “ u (15.-13901 . , (3.9: ting-n 5 (131/ii 1090291 ‘L brim '9» .318 €- bns 9T Io mm m amoia .15 24:32.0 vllsoidqmpollstavto owl 616 919rlT ,uO 10109100129138 “ \(lfrlgtle 330 691.1 [3(236TmuD) 90.1 m mots ( :in 5411mm W” egmevs MT .11 0mm: 301016108 tunl rlfiw inremoeg W” A" 3‘»): ‘1 269 ....... Figure 60. ORTEP representation and labeling scheme of the unit cell of CU17.6T98328- *: lleo IIRL M1 to emertoe gniiede ”has "13118208391991 RETRO .03“ 6:36“qu 270 Cu(1)-S(1) bond distance is 2.333(3) A which is comparable to that in the known Cu12Sb4S13113. The Cu(2) atom of the CU2,BS unit has trigonal planar geometry with one S(2) atom and two S(1) atoms around it. The average Cu(2)-S bond distance at 2.182(6)A is slightly smaller than that (226(2) A) in the known Cu12Sb4813, but is very similar to that (2.192(4) A) of three coordinated Cu atoms in CuSGZ. The coordination environments of the metal and S atoms are shown in Figure 61. The Te atoms are situated on a 3m crystallographic site, possessing trigonal pyramidal geometry with three sulfur atoms and one lone pair of electrons. The lone pairs of the Te atoms in the {Cu1z(T683)3} cage are related with each other by a four fold screw axis. Four of them are pointing outside the cage and the rest are pointing inside the cage. The Te-S bond distance at 2.377(4) A compares favorably to that (2.358(5) A) of known BaTe83114. Without considering the Cu(2) site vacancy, the S(1) atom of the {Cu12(Te83)3} cage has tetrahedral geometry with two Cu(1) atoms, one Cu(2) atom, and one Te atom around it, while the S(2) atom (on the -43m site) inside the cage has octahedral geometry with 60u(2) atoms around it. The Cu(2)-S(2) distance is 2.189(9) A. Selected bond distances and angles are given in Table 86. l “ i: "U l.- 'l t v 91).. if I: seem 3111’ ft Mums em ("8 “In “in w (A (3)85: 2 mm nerff tolhmz vlmtail‘e of Am: in wt: to 1'5. Inger 9) mm «:1 WBl‘fltla m am ...e. . '~ 1' .l ...-1 ta t"‘=?$f‘lh1f.03 911T 981.10 fit one. I0 *, ~ t " 39’ 1‘3 mum"! ni rtworle “I“ . .4 : t, .: Lttaer-eeoq .ells ., m... . I 3.2.6.2116 o ‘. .sq enol one MI“ I“ ~, :3i6‘ s ‘3 99.63 18188.17!” .‘ o gnifftloq 9‘ M b a. F, a) r~=~a=h mod 8-6TIrfT I l tf :51-12. .1. - v - J. t 7 " grfieTea cm I V" 621283719200} Id! b 2.10 DOS ,mofI (SM . 9.1 6060 0111 ebIII'II,‘ 24.06180 (SE-(9)00 :i-zal eldBT at m 271 (1 Cu(1) ..- V] S(1) _J W i A O O S(1) (C) Cu(2) (D) S(2) (E) Figure 61. Coordination environments of (A) Cu(1) atom, (B) Cu(2) atom, (C) Te atom, (D) S(1) atom, and (E) S(2) atom in Cu17,5T63S26. mots (94.0 ('8) mos (1)03 (A) to memnmivne 1101mm .fO “ .ageaeTaxtuO m mom (9)8 (‘3) hits mats (1)3 (0) ”I? (a ‘ ‘0; , . * 7' 4‘ .- '1 . 272 Table 86. Selected Bond Distances (A) and Angles (deg) in Cu17,5Te3$25 with Standard Deviations in Parentheses Te-S(1) 2.379(4) (x3) S(1)-Te-S(1) 97.5(1) (x3) Cu(1)-S(1) 2.332(2) (x4) S(1)-Cu(1)-S(1) 111.93(9) (x2) (1 )-Cu(1)-S(1) 104.7(2) (x2) A“ S(1)-Cu(1)-S(1) 111.93(9) (x2) Cu(2)-S(1) 2.171(5) (x2)S S(1)-Cu(2)-S(f) 95.3(4) L Cu(2)-S(2) 2.201(7) S(1)-Cu(2)-S(2) 132.4(2) (x2) Te-S(1)-Cu(1) 100.6(1) (x2) Te-S(1)-Cu(2) 1 12.7(2) Cu(1)-S(1)-Cu(1) 101.6(1) Cu(1)-S(1)-Cu(2) 119.1(1) (x2) Cu(2)-S(2)-Cu(2) 90.00 (x12) Cu(2)-S(2)-Cu(2) 160.00 (x4) a - . h warm: . M. l (l)8-( MU ‘l; 45,” ..91 3 2 .1 I ‘5‘ '. o , m mom" I l " .9 ti ( (‘11) ‘i, ‘v _ __ -_—-v———--—--,-* (nest-l r18 ' "\JIJ‘t‘ (‘1 O-(l)8 273 3.3. Charge Transport Properties of KCu482Te (I) Based on the band filling arguments advanced above, KCU4S2Te and K30U3S4Te2 are expected to be p-type metallic conductors. Charge transport measurements over the temperature range 5-300 K on single crystals of KCU4S2Te along the (010) plane show that the resistivity at first decreases linearly with decreasing temperature, but at low temperatures levels off to a constant value (so called residual resistivity) as shown in Figure 62. The resistivity increases from 1.5x10'5 Ocm at 5 K to 1.2x10'4 0cm at room temperature. These values compared favorably with those of KCU483 which are 1.67 x10'5 0cm at 20 K and 2.5x10'4 0cm at room temperature. Even though KCu4S3 and KCu4S2Te have the same number of charge carriers KCU4S2Te might be expected to have increased conductivity due to the increased overlap between valence orbitals of the larger Te atoms. However, the similar conductivity values clearly suggest that substitution of Te for S has very little effect on the band structure. The temperature dependence of the thermoelectric power (Seebeck coefficient) shows a very small positive value of 1~3 pV/K in the temperature range of 50~300 K as shown in Figure 63. The small and linearly increasing Seebeck coefficient with rising temperature indicates that KCu4S2Te is a p-type metal. Unfortunately, charge transport measurements on IQCusS4Te2 are not available yet. They are currently under investigation. noqanstt 9916110 .2th aim ‘1‘ lo aisfa )3 sigma no )1 008-51 99081 M brie skews» m. i “ .L. -.-‘ * r. '21:? is wit/1192': :1 91f 213312 wode em (0763“ . t . 75'.» .. 31.:129 w.‘ ‘3 :ud 91013131311010!“ :‘ .9 - 6361...: belies as) out- 1 ,1 '1 .. Jhil‘ moi? 388mm . ‘1'." ‘ Jo ’ '3" Am. “,' ’si lgv OBSNT .M 1 , '_ 9.“: -‘ c. 7 w is 9.10 3‘0” “id ,_ . ,1 _- 4- . ’h: .1,‘ $.05 8830331 1“ 1 > - "‘9 “'Q‘t‘.) 0733.8” . r- ...t ._ - ',l ‘ ‘11»:«1‘159nsvom 1, _ .: . . n-Itwtouonoo telimb If. . an: no 130110 I”!!! w . ,7. ~v, 4.....2— txsuuvemsqohm \i (j, . . ‘- 'f ‘ ..tr'" (stints? 9111 m NV“ ‘4 "1:11 \ghsenll MI in 7., . .- . CsftdePl mmm r1111, . .. ». ' 97+ rc. l _. a: 3,":t9d" .vletanunotnU ~: . r _ .9 a 616. vertT to!” . ... 274 ....n!'. 945" ..E I 093.59. .o .2298 065m 6 .2 2288.5. .o 20:82 a as San AEoCS 3257.0. 393 So“. do 2:9... 0: mSfiEaEmP com omm com . omp cop om o d W q q — a q d — - m u q — u d u q i q d u u _ - a u q o (IL—0.011) All/\llslsea . tom? 03 ....w. m... 7 .. .11.”. ct he.» .9 5. 8921216613 l ifngUJ) 27S .ohmwvaox .0 .225 06% a .2 53 0.35 338 22.00.005.05 05.8852-o_am_3> .8 9:9“. 0: EBEmanP 0mm com omm com amp 2: on o -«nu—quuq—uq-q—d-qd—uqqu—d-gd—qu- oI—u .1. .msm. . w 1 O - w loam . m . .m/ - M 13m od 1‘ 5:9, :2, ' Ci "V .1.“ 1!. wobcmet (ti/(\K) a in.“ 2““ 276 3.4. Magnetlc Susceptibility of KCU4S2Te (I), K3Cu3$4Te2 (II) and CU17,5T93825 (III) Variable temperature magnetic susceptibility data for KCU4S2Te and K3Cuas4Te2 are shown in Figures 64 and 65. They both show similar behavior; (a) temperature-independent paramagnetism above 50 K and (b) rapid increase in susceptibility below 50 K. When plotted as X" vs temperature, for (l) the data below 10 K follow a Curie-Weiss Law and for (II) the data below 30 K follow a Curie-Weiss Law. This behavior is characteristic of metals (Pauli paramagnetism) containing a low concentration of paramagnetic impurities. The temperature independent Pauli paramagnetism ()(,,=9.7x10‘5 emu/mol for (I); xp=30.5x10'5 emu/mol for (ID) have been obtained by subtracting the Curie-Weiss portion of the paramagnetism from the measured magnetic susceptibility. This Pauli-like behavior of (II) is quite different from the isostructural compound K3CUaSe which is known to have resistivity and magnetic susceptibility anomalies associated with charge density wave (CDW) states71lb).(°). The CDW or spin density wave (SDW) state is one of the interesting physical properties observed in low-dimensional metals and is associated with the electronic instabilities (electron-electron repulsion).77 A recent theoretical study115 on IQCU3$5 suggests that its observed anomalies do not originate from CDW instability, but are caused mostly by an order- disorder transition involving the Cu+ ions of the energetically unfavorable long Cu-S bond (2.84 A). Cu atoms move back and forth along this long Cu-S bond to shorten or lengthen Cu-Cu and Cu-S distances. These superlattice modulations and resistivity anomalies are not expected to occur in the higher chalcogenide analogues of A3CU335 because they have more reasonable Cu-Q (Q=Se, Te) bond distances. In fact, the substitution of Te for S, which “In 3r...“ “'4 ' 61‘ bus sTgesUOX 10f 8390 ”W w “131111118 nvortze rllod 99.17.66 has #8 w '0 fl on: .4 ..1. 3"»:st masngsmmsq 109W - ..n, ens! P! 66 misc vl‘ilidltqeoeua N 4. 9:». u . .w’ll’.’ >401: voted ntsbertulhd 7 l 2‘- ‘ JEJ E’ .Q‘N 511110 3 «:00de 08 ' . ...l c ‘....-:.-:....at19606rnsteq M 1 , it"; '1? 111.9: " ...?“ 7' ttxsteqz'nef 911T v.4 ; 4.3.16.9,- ,(l) 101 1‘71""! .' , '“c: z ‘i "-c t‘ ..‘f VFW") 9’" . ;t... 4,4 ..nn. {‘- . 11‘:",BO€U¢ w . . . ,1 -\,. n) T (...-g" . P11011100 lstutatnbed - - '. v. .‘.“."!Ci » :vliuifqeosmw 4‘ 7 . w , _ _‘ h .1. '0 ,,. 3.412 mum...“ ~~ 1" . sew“ ~ 1. 1.43:6de WIM‘H -._‘ '4. g. .r we. srll rtfiw be!“ _ .- ..wht rt: “acute lsoifetoorff M T" M, _l ..r‘,.=:.-“:Ia." ‘1'!) 1.) H1011 BIBMQflO ton d 21".: DE'euE’) a «- grin: yln‘;t:,\\v.rf7.‘ "7';'.‘lh.’-‘,~';i'l"f9 4.5"] la: .5113. ‘9.) 3'11 p'tiv'zovnl 11013180313 M‘ _ , , _ ,. \ Drzccl €64.72) 131'” 1 -;’l.’" t‘lmit-z 'llit'v' fists 4.39:1 991:1; {I‘ll-S 1k) (A 88.8) DMZ-d, 2330351161368 1326."? 690.033.35- 2.9.1.“? brie. 01’; normal 10 hem I 190ng ad! ni tun-'3 of 06332:: «.2» Ion 616 aatlsmons vthfsieet Ono am eidsnoese‘. 610m evsrl vertf 9206090 38300;}. to 2609013119 cum (fetttw .8 101 of to noitutilzdua em .1961 nl 2991161610 bnocl (eT .030) M 277 H 9 353.25 ...:x ”tome. .ohuwvaox o£=wfibozon .2 83 205358 3.33888 0:959: 229an2¢3¢§> .3 2:9". O: p com CON cop o F . . . O; a a a a a a a a a a a a a . —..—. IN; c: h . fl . Z c. m a s m m . .m. W . . . p . . p . . p . $0.0 I 4 W m... “4: m a [‘0 \I a w a 4 M Bum-F “u H Imam m H a .4..pr .m P a NS 3 A7650 l\ P .23 o a . . ' vim. ‘. M A.” _. .u 4n. m. u .. U of: . . . .r ) Pu m. m I . .5 w . w .. Hi3-.-» J r m T: a La .W. 1:!le 1.4 -~ 278 ... m> 369.25 7:2 “:35 .uchvwoaof o:_._£mao>_oa .2 22. 206383 55.3083 022.92: SBSanoTontg .mm 059". 3.: com com 2: o .[ . p o r I I a I I I I I I IIBBIIG anae- aa :— a. a, cap a 3 cm 3 c. o a-~ W . u b p . G Am. and W. I m $.49 . ).. -m w _ m a mu . . m a (W 1.? I Acts , Na E‘Ambtc m ""515: . .,-- _ 4.. xu‘ (mmm) 279 replaces the quite long Cu-S bond with reasonable Cu-Te bond (2.97 A), does not allow any superlattice modulations or anomalies to occur in (ll). Thus, (ll) does not show any magnetic susceptibility anomalies, but shows Pauli-like. behavior of a typical metallic conductor. The magnetic susceptibility data for Cu17,3Te3826 are shown in Figure 66. They show antiferromagnetic behavior and quite strong magnetization of the materials. Because the magnetization shows field dependence we chose a magnetic field of 100 G; this field gives a valid linear field dependence. Based on the stoichiometry of Cu17,6Te3826, we could expect a maximum of 3 Cu"2 ions per formula unit. We were expecting weak paramagnetic behavior of this material since the known tetrahedrite compound Cu128b4S12,7 exhibits paramagnetic properties due to the localized Cu2+ sites.116 However, the magnetic moment calculated at 300 K is 33 BM which is almost equivalent to 32 unpaired electrons per formula unit. Figure 67 shows the temperature dependence of the magnetic moments over the temperature range from 2 K to 300 K. Magnetic moments increase from 2.4 B.M. at 2K to 33 BM. at 300 K. We are not quite sure of the origin of this stage. First. we suspected possible impurities in the sample but the X-ray powder diffraction pattern measured on the same sample showed only the pure Cu17,5Te3826. Further work is necessary to understand the magnetic properties of this material. «atrium (il) .etufT .(u‘) now“ sun-nurse: ewode lud .aellsm M {at} m (“Jada 335 8:891:13 {[113 micron m ~ 4'er one towsried aim ‘- ‘ . ”"fia-L: are“ swung ncszasatengnm or” m .,_‘.._.,.‘-44 i 'i“_'. .‘l tilt”! U“. 3 ‘ ‘l s ' l i al'i ) ,i o. col \ i ‘. “ 1 . l, \‘ " ,1; . 4 --:-r m y-frTa'HLL) to mmm i ' _ 'r- 2 ‘- oneqmq sitcom ‘00") Q J Ham“! 3'1 I a a" 223,5": b 35%;; 16) 31:1? D 00? '0 hhl- '3 V r 3d r;r-.’;.s-c,. 95w PW Pint: alum“ , I 1‘. ‘1‘”; f‘l'v{"‘l. :52, 90”” “- .‘- w ~ ~ ' is rv-.s.l.;:-;imemomw . ‘ i . '; .3 l)"l'“l" ' WC! .'.'iJ1’I)919 m1!“ 1 m ' =.‘.}' .' il‘i‘bh'l or?, if) 8mm .- cizfengsM Jim; R if: Hue eflup for: "I .. ,- , 43 emmsa ed! ni W . ll. . .7; :1 “my: mung”; glqmsg 3m 4 s: .; am. y :iswgzv 4.:35fa‘lebnu of m D 2‘4J4 \V \ 280 tan emmoohofiao o:___2m>.o>.oa .2 San 208382 £53083 0:252: o.:.a.oqE2-2na_.a> do 0.29". CE com CON on: o . m.m Wad I . I new an -N.v a . L; a . A725 .34 (low/hula) wx s .- ".~ biffik '5‘ 1', -‘,v - " (Hi?! rm W'lrrlfl 3‘.“ ll ._ I l U" L‘. 281 .oumoOkoNpao .0 A29 29:2... 2882. 38.3.3 .0 859.83 2323th No 2:9". com CON 9: GOP an .3 .ON .94». O¢ (we) tuamow onaufiew .‘y" 23w} . v;1,1.\ V _.\ 4 f. 4‘ ‘4. ‘ ’5) -1 ‘ 1.. (s: 1 _L) ‘ l l k I Q .1 I , l ”:3; '\ .E‘c.’ 4. --.- -_... - _.. -- f v ‘i ‘1‘ ‘4' O 4 L7 01 282 The investigation of mixed chalcogenide compounds in K28/T e and K2Te/S fluxes yielded quite promising results depending on the flux compositions and reaction temperatures. The compounds of KCU482Te and K30u8S4Te2 showed one important feature; Te atoms can have a preferential coordination environment over smaller sulfur atoms in a solid state lattice. This feature may make it possible to stabilize new structure-types which may not be possible otherwise. The compound of Cu17,6Te3825 highlights the potential use of the ['l'e8312' unit as bridging ligands to stabilize extended solid compounds in the mixed-chalcogenide flux (AzTe/S). The [Te8312' unit, formed in the sulfur-rich flux, may be a potential ligand to use for the synthesis of new structural compounds. Since we have investigated only K28/T e and K2Te/S mixed chalcogenide fluxes, there is certainly a great deal of exploration remaining for new mixed-chalcogenide compounds with other alkali metal (i.e. Na, Cs, Fib) Te/S fluxes. geneauogx one 91M comm :W ‘, noltsnlbiooo lsifne1et91q s evsrl nan ”I? w W ‘(sm emlnel zirtT .soiitsl stale biloe s mom-m ":Lflrf.r3l".';} :1. m1 vsm rloirtw aqu-smlomu won as” or 1;" {my a: .'z“.“.':;’.‘. vrff alrigjwlrig: '1 egaaeTa‘nuC) l0 mm I" . > . '_ "1:; M 3:. 16.33.01? .5 whiff-33?. 03 8mg“ 9mm 8.“ ‘ , . .vj; {‘f MT {Fifi-SA'Ixult ebl . .mr-L‘ an (.2 bung“ lsitiiefoq I .6 ll. * .. L. .ng. ' r-kir-www 915.1 ow 50(13li ‘ ..Ji _ , _; . :,'.f:";“l.'?1c~"l esxu" ohm . '. '. . .. ' 1...; rv‘- ' '..:‘..|‘qr1' finegoolsdomm . .aoxullfl' P CHAPTER 8 AuCuSeg, a Novel Mlxed Metal Chalcogenlde Compound Incorporatlng $032- Ilgands 1. Introduction We have demonstrated in the previous chapters (see chapters 2 and 3) that by using alkali metal polychalcogenide fluxes as solvents and reagents at intermediate temperatures, novel, low-dimensional polychalcogenide compounds in ternary AlCu/Q and A/AuIQ systems can be isolated in crystalline form. Based on the success in these ternary systems, we have extend our work into the quaternary A/Au/Cu/Q system with the expectation of new polychalcogenide structural compounds. There is only one known chalcogenide compound containing Au and Cu metal, AuCuTe4.117 This is known as kostovite mineral and contains Te22‘ ligands. There are no known mixed metal polychalcogenide compounds containing higher polychalcogenide ligands (0x23 x > 2). Prior to our work there were no general synthesis methods for polychalcogenide compounds. In this chapter, we illustrate that our initial approach, using alkali metal polychalcogenide fluxes with mixed Cu and Au metal at 310 OC, leads to the synthesis and crystal growth of the new ternary mixed metal compound of 283 y\‘ ehnepil $502 autumnal ‘19" 5533.3!" ’ . up. ‘: a 3‘ \“'v. — .4 ., ...: .r: w hamrfimmeb WW' . . r' . 3.: '23‘ "gficw ‘3‘”2~.1ll“e/‘filmb W“ w ---~m=“aqn19f slamm- .'d .1- ,. .I r, 1 v" A hit. 1).: -4! {camel m M .‘ , 3 ,—.- .F H; 3,41.” l' ,;J.'~'\0U€ 9:11:10 M u .7 _. ... _y u...’ .; 1n. vvcfielsup 0d! “ .1 ,., 4;, ~. ,:;1L"v":: eblnspow r, ‘ _ smut b-“uU-Ql‘nO‘.) 9W i g - i- 1:; .‘r mmm sfivofaox BI M . ,. e , -. .. .::.~.v -- '-«~~.w:wladovloq W w ’ an L W". “4'; 2,3;ie-fiz‘ix‘ilj' 01 232:)". New .~ "Jr: m“. ")3 "7‘53 l3 < x 3‘10)“ :nmw‘yt'oo etmeW" latem ilsxls mid!) ,rlosmqqs iszflm 111:,» ism edgingulli em .16me M II art} 0! absnl 00 NB ,5 gsfem L1,“. bub rt) mmm ri‘nw eexuli shim 10 gnuoqmog lefeen bexim wants: wen. em to mwmg lsfa‘no bi" w 284 AuCuSe4, containing bridging triselenide ligands with a novel three- dimensional structure. 2. Experlmental Section 2.1 Reagents Chemicals were used as obtained: gold powder, -325 mesh, 99.95 % purity, Cerac, Milwaukee, WI; copper powder, electrolytic dust, purified, Fisher Scientific 00., Fair Lawn, NJ; selenium powder, ~100 mesh, 99.95 % purity, Aldrich Chemical 00., Milwaukee, Wl; potassium metal, analytical reagent, Mallinckrodt lnc., Paris, KY. 2.2. Physical Measurements FT -lFi spectrum of AuCuSe4 was measured as a pellet in a Csl matrix. The sample was ground with dry Csl into a fine powder and a pressure of about 6 tons was applied to the mixture to make a translucent pellet. The spectra was recorded in the far IR region (600 to 100 cm'1) with the use of a Nicolet 740 FT- lFi spectrometer. Quantitative microprobe analysis was performed on a Jeol 350F scanning electron microscopy equipped with Tracor Nothern TN5500 X-ray microanalysis attachment. Single crystals of AuCuSe4 were carefully picked and mounted on an aluminum stub with conducting silver paint to help dissipate charges that developed on the sample surface during measurements. Energy Dispersive Spectra (EDS) were obtained using the following experimental set- up: .rl . ,5. ‘.)‘_ 1; f_ -. ,lfi""‘ A .y‘ lm") 1W .fiefluaww . . ' r !~ ‘ ‘t' 1”” V’- " “*6 1M meal 138:1..w .’ I . '4 x , .. ‘. I r “ . .~- , , ”we so laoimerd . ‘I -. in .te-nsofrio as beau new o "H .4157! . onl ..L ..ieit'rfirl‘? "lC-ieYflq .u Hill“. . ”w a r , "m “who . l’. rxtvtroaqa Fit-n“ ..‘JW .- , . _.-“ ‘1 “Eu Mums; saw 0”“ .51 -, 4" .' 7 ...-hr gulf 03 Comm” . at 3:13.], ”giga- iTtl 1810!” M m M ' - 1.. «an. e m 'QlTilCE’YFOQ agar. gzzrevilme. edenqotolm sv'uafifwgw vs". a 006d!” marital/l toos'iT rifle; i-rBCg"‘.:..'l‘.'-.' .uosaomm flannel. m beflolq (imitated 319w ..aEUOUA lo aisle”:- atom}: themrlosng m elmiaezb qleit of leisq 1evlia grufoubnov nfiw (tux: mummuls no new _ mien-3 amemetuasem gnhub soshue slqmse srlf no beqoleveb mm in lsrnomneqxe pniwolloi orii gnseu bsnuszdo mm (803) M m: . .‘u 28S X-ray detector position : 55 mm Working distance : 39 mm Accelerating voltage :20 KV Take-off angle : 27 deg Beam current : 200 picoamps Accumulation time : 100 seconds Window : Be A standardless quantitative analysis (SQ) program was used to analyze the X- ray spectra obtained. Since the selenium ratio is always underestimated due to artifact of the program, a correction factor (x1.86), which was determined by calibrating with known KIAu/Se and KICuISe ternary compounds, was used to evaluate the selenium percentage better . The analysis reported is the average of four individual measurements on different crystals. 2.3. Synthesls Chemicals were measured and loaded in Pyrex tubes under a dry nitrogen atmosphere in a Vacuum Atmospheres Dri-Lab glovebox. Potassium monoselenide (KZSe) was prepared in liquid ammonia from potassium metal and elemental selenium in a 2:1 ratio. Copper gold tetraselenide, AuCuSe4 (I) 0.094 g (0.6 mmol) of Kgse, 0.098 g (0.5 mmol) of Au powder, 0.032 g (0.5 mmol) of Cu, and 0.624 g (8.0 mmol) of Se powder were mixed together and loaded in a Pyrex tube which was flame-sealed under vacuum (~10‘3 torr). The tube was was placed in a computer-controlled furnace and heated at 310 0C for 96 hrs and cooled slowly to 120 0C at a rate of 2 °CIhr, then to 50 °C at 35 °CIhr. Dark brown rod-shaped crystals were obtained with contamination (about 20 °/o) of KAuSe5 by removing 1"?" I r A " .. h ' . ~ ”5., h: 1 emit v Wm ‘ f. sl- “. 5.1-ink '7‘ 10 (1.9) em ”awn” 3. 3492 Mt some M" . ' "WV ‘u‘ 64‘ :1. 1' 14.0mm 9111:1000 «It: ‘2' ehnooee 00f : * r .2 v: rr. »‘ :3?" ".5 h... ‘P‘uA‘Q‘i nwoml rftiw . . “c: 1' ‘ l. , " 3:; - .35308316'4 muonelae ed! 3 , ":9 mm“. m 1~>.;7r39rnmueeem leublvlbri aisedfnya £3 3 '4 r >~ ..-:;'-.3c1 Stew 218$me .> 4.,_ ,; , -. ~ iii-4:2"; -: ~.. aiedqeomfa m .. 3 (1"“1” .; . . . >. .;:n‘:‘r'::5'-- '3'. «55w 63.9).) Obiflw : 7 1 -.-‘. fl' muineise [3me“ a} ’0 lionvm t} '3" _, " 3:. x) (2113.92.03.41 .shifisinsmfe! biog “an: 9 #88 0 MB to (lomm 6.0) L, Ed) 0 flat my, 0A to itclnm 8 O) 0 “40 fl naifw “do! x51.“ 5.. .1: bebsol bras write-go; mmm 315w iebwoq 98 tom 5 m booslq agsw 25w edul 9!” .(11-1' Jain-j) mumsv what: how w M belooc W 31d 69 1m 30 ore f5 belsed one eonm'ui Walnut-m male-em maid msO .unou as is ou ca oi hem mac 5 to ennui?” W1 yd 9,98qu to (JP OS focus) nonsnimsfnoo rffvw Denim m - , 286 excess potassium polyselenides with DMF under a N2 atmosphere. A quantitative analysis performed on a number of crystals with the EDSISEM gave an average composition of Au1,oCu1,oSe4,1. This compound is insoluble in all common organic solvents and stable with respect to hydrolysis and air oxidation for several days. 2.3. x-ray Crystallographlc Studies AuCuSe4 was examined by X-ray powder diffraction for phase characterization and identification. The d-spacings were obtained from powder patterns recorded on a Phillips XRG-aooo computer-controlled powder diffractometer, operating at 40KV, 35 mA. Graphite monochromated Cu radiation was used. The d-spacings observed for the manually selected single crystals of AuCuSe4 were compared, and found to be in accord, with those calculated from the single crystal X-ray structure analysis data. The calculation of d-spacings was performed using the POWDfO programs“. The result is summarized in Table 87. A W . . ' “tin-if M36803 on: m mo 19 ‘ alduioeni 2i bnuoqmoo am .MOH.” “E. has ?‘.2~.Ilmbvrf of magnet m m in .J‘I‘A ‘ ‘ . I ‘ 5’ o. . . masque olnqnponmuo m w} , 7 l fin“ . .4 -. ;>. Hi _ .. , “ V'l-flw °6w 50839096. I. -, y 3. am‘ 5.,‘181‘iflf‘3hl bus RM . ' yr. ’> = 5. ‘c‘jw' l? e; -- bemoan on.” f“ “I 0. l)"‘i‘-.'.:IQO 1919me . , , 35¢ 1114.. ..;“‘::uaswnm ~. .~.. ,.._ .._~,.v>:— - w >.?,..,;)l.vA to am '3 .- p ..2 ”3‘41 mm? mm M . - " 4. a ~1>q ”8w egnioem-b N J 5' 1a? ni mmm 287 Table 87. Calculated and Observed X-ray Powder Diffraction Patterns of AuCuSe4 H K L 8mm dobsoi) l/lmax(obs.) 0 0 1 8.08 8.09 14.1 -1 0 1 8.89 8.90 29.5 E”? 1 1 0 8.78 8.78 15.1 I 0 2 1 8.88 8.88 14.5 ‘-- 1 0 2 2.88 2.87 15.8 -; 1 2 0 2.81 2.81 100 L. 0 0 3 2.69 2.69 63.9 -1 2 2 2.354 2.858 8.9 1 0 8 2.227 2.229 8.9 2 0 0 2.155 2.155 25.9 -1 2 8 1.985 1.987 46.8 1 2 8 1.9117 1.9128 87.8 -1 0 4(0 4 0) 1.8724 1.8889 82.8 -2 2 1 1.8879 1.8888 29.0 1 0 4 1.7905 1.7908 12.0 0 2 4 1.7784 1.7785 26.7 -2 0 8 1.7880 1.7848 18.4 1 4 0 1.7091 1.7100 6.1 -1 4 1 1.6800 1.8787 11.1 2 2 2 1.6622 1.6629 80.8 2 0 8 1.6375 1.6380 28.8 sci. 08¢ . ".13 S S W e 0 nwo' 0 0 9 "‘0': 8 S I‘- ' it"? 8 S 1 tall 1'- :1: a 0): 0 r- rss- ior 930 609- oer ref- 3:: cos 288 The X-ray single crystal data of AuCuSe4 were collected on a Rigaku AFCGS diffractometer with graphite monochromated Mo-Ka radiation using the 011-26 scan mode. Accurate unit cell parameters were obtained from the least-squares refinement of the 26, u), x, and 0 values of 25 machine-centered reflections. The stability of the experimental setup and crystal integrity were monitored by measuring three standard reflections periodically (every 100 reflections) during data collection. The intensities did not show any appreciable decay. Two absorption corrections were applied to the data. The first was an empirical absorption correction based on 111 scans for three reflections; this was followed by DIFABS57 correction. The structure of AuCuSe4 was solved with direct methods using SHELXS-8653 and was refined with the TEXSAN60 package of crystallographic programs on a VAXstation 3100 computer. All atoms were refined anisotropically. The complete data collection parameters and details of the structure solution and refinement are given in Table 88. The final atomic coordinates, temperature factors and their estimated standard deviations are shown in Table 89. . _I' "'.1 .' I' 1’ I... ‘ ” :J . _. .- 9333111. eefsupa-WM'UMG‘W ' . .1..- enT 211011081191 momentum: 318W .x nemtenom 915w 101199nt Islam M m . 1 ‘ 1 (”aru- .1 git-:1 0:51: 112,883 vilsoibo'naq 8W m ' t; .“ .t k 3 833‘ ‘ o 3“." . ., . "J x " 9. “. L ‘ m; of patterns 815w anoffofllu' 4' -...r -..-_, 1:123:12 11.1 mb earllanolni MT . ‘ 1 3“L,¢' ,1 .. . "rev-8 111.218.25.an ‘ ' . . Ll . 1 118-1: rm 000891100 M‘Id‘d' _. 5 ‘11,, , - ,wéa, . flu: abs-{A 2T'5-'?H8 uni-1.1 M 1 .mr. _ , .r ‘2.“ .1” u no an:..t:-O“.q 3W r) , -, g ‘1. ‘- .5.:.;.‘;_-,(. -_- ' $5.71an”! ;. ,1"?:: DW- 1101:0108 mm a , y, . _ 4r we 2"?i-.).-' ‘-=T7-..8’l33m9, 3.18m 98 alarm” 289 Table 88. Summary of Crystallographic Data for AuCuSe4 mmmund l Formula AuCuSe4 Formula weight 576.35 space group P21/m a (A) 4.818(2) b (A) 7.447(1) c (A) 8.099(1) a(deg) 90.0 8(deg) 9333(2) y(deg) 90.0 Vol (A3), 2 260.0(1), 2 Temperature (°C) 23 Crystal size (mm) O.60x0.05x0.05 Radiation Mo-Ka u (Mo-K01, cm'1) 597.8 Dcalc (g/cm3) 7.36 29max (deg) 60 Scan method w/26 No. of data collected 1698 No. of unique data 815 No. of data used 623 (F02>30(F02)) No. of atoms 5 No. of variables 35 Phasing technique Direct methods Final Ft/FtW (%) 5.1/5.5 Max. shift/esd 0.00 (last cycle) Extinction coefficient 1.01x10'5 \ 1 I i .___ _... ‘J r ”(1,. ‘ s .1“)va -”."°)o1umoqu m- r'u“) axle M ‘ noifsibofl 5' '1'11." .1 NM 1' 5 " ("-v'f'lJ‘tQ} M- . . Q90) ”MOS 38.1mm r1808 r‘ .1100 81.0th 0“ v0 eupinu to .old "4350' efsb to .CM (1501):;8601) emote to on aeldsnsv to .oH euplndoot GM (81") "MR W bae‘mirfa .10“ (aim in!) 73181353301) mm 290 Table 89. Fractional Atomic Coordinates and 89q Values for AuCuSe4 with Their Estimated Standard Deviations in Parentheses Atom x y z Beqa, A2 AU 1/2 0 0 060(3) 36(1) 0.4162(3) 0.0013(2) 0.6951(2) 080(5) 38(2) 0.0742(4) 1/4 0.6384(2) 086(6) 36(3) 0.1445(4) 1/4 1 0623(2) 066(6) CU -0 0248(6) 1/4 1 8423(3) 1 .5(1) 3 B values for anisotropically refined atoms are given in the form of the isotropic equivalent displacement parameter defined as ng = (4/3)[a2811 + sz22 + 02533 + ab(cos 10812 + ac(cos (3)813 + bc(oos a)stl. 4 if '2 Hr.- | a _..)q: .- - 1 18181000 (mm 11.‘ ‘_ ' Ml’ (MSM’OD i1 1 .- 1 (8)88800- «r—I ' ‘2‘. 1 1"! . "I 1.- .' Showoelns Mouthww 1 .-- .1 «aren’tslqelb 1118mm. . f';(~:f‘. ; ~'??l\'300h8+$%;§ I ‘. .l |’ I q .va {.11.}! ~. ' 291 3. Results and Discussion 3.1. Synthesis and Spectroscopy Synthesis of the novel mixed metal polychalcogenide compound AuCuSe4 has been readily achieved using potassium polyselenide (K2Sex) fluxes as solvents and reagents at 310 00 as shown in eq 1. 0.6K2Se + 0.5Au + 0.5Cu + BSe ----‘-> AuCuSe4 + K2Sex eq 1. Our initial approach was to use reactant ratios of K2SeICu/AulSe (from 1/O.5IO.5/8 to 4IO.5/0.5/8) at 290 00. Upon keeping the sum of the metal ratios as 1, the overall reactant ratios were same as those used in ternary AlCu(or Au)/Q systems. During this study we identified at least two new phases, including AuCuSe4, that were competing with ternary polychalcogenide compounds (e.g. KAuSe250, KAuSe553(°), a-KCuSe450). The presence of the competing ternary phases was not surprising because each metal has its own favorable coordination environment associated with the various polychalcogenide species in equilibrium in the K2Sex flux. Microcrystals of AuCuSe4 were obtained at 290 0C with contamination of KAuSe5 as a minor phase from every reactant ratio except that of 2/0.5/O.5l8 The product looked like a coarse powder to the naked eye. Since the ratio of 1/0.5/O.5l8 gave slightly better crystalline material, we narrowed the search to reactant ratios around 1/0.5/0.5/8 and increased the reaction temperature to 310 0C to improve crystallinity. The O.6IO.5/0.5I8 ratio gave single crystals of AuCuSe4 suitable for X-ray single crystal study. However, the product was still contaminated with KAuSes. Another promising phase was isolated from the DDUOU'TZOI‘. ebmepoolsrfovloq latent m M . 4:93:11 -;_1t1.:u:alea\rloq mueesfoq (70180 M M - t 5311"“.01‘43 .35 3001818 WU. ,fih-Qy .-: -» 826+ 038.0 + uAa.0+ ~ ... . r ‘ _.. .Acx,‘ 82L «81' '1oso'qqs ismw ‘ 7 ' ‘ 1.1‘ 1' (1., ‘ 1‘ _1 . )1) 33- 2-1.3 ’9‘30‘30“ 01W . . , . J. . 1 3 ..11158811181ovouflfi. .3 . ' '09: : 1' 9111100 211193“. m 139 isr‘: ;.£18UOUA 9M 11. *1 .1 "-1.; buA}! '99) am >.,_ ...g -. '. ‘2 '11 e. man-1Q 11181716! 90*" .. ..uw ». «51101000 old.“ '1. ~ '. 331.7882 561mm . 1.1:. . «33.1/I 10 81638‘000191“ . v .> *3 1".)f'ujfl-3 g ,4 1(3-5T3» _ ; ‘w: 31".”) $128110 iC’YimB 33M ~. '5 SW: rv‘wsx. :51 1 3 :1 I“-'7::011 t. 91111 0311001 m .17 13.7 ‘-7,::";—a ‘11'.’ 5"3'2.::-:nee'. 3W intake" Ef;‘.‘.!lé.‘_y'iii 13118(1 113109112 W89 “Inn? 01 mum-101.117.1142. "0178391 3:11 beeswax» ":1: on: .113 OH bmwts eoflm W1 "’ czleéav 1:1 91:31” 5118;; 0125'. 9.6018 {116.0 “n" 1' vflfiillsfev‘m 9W 013° 0" K] :»~ W. 2818 "11.5011; 9111 19179on vbuta 1.8181119 919012 (SAX 101 W W 311? mcw'r 89181081 88111 828110 0111-2101010 :s-rffanA peBuAX rfflw W 7' :' 1:53:34? 4" 292 reactant ratio 2/0.5/0.518 at 290 00 along with a-KCuSe4 and KAuSe2 both of which were identified by X-ray powder diffraction. This phase has the approximate composition of KgAuzCuasezo by EDSISEM quantitative analysis. However, its very poor crystallinity and very small, thin needle-like morphology made further characterization difficult. Further investigations to obtain better single crystals of this quaternary phase are under way. In the far-IR region AuCuSe4 exhibits spectral absorptions in the range of 170~235 cm'1 due to Se-Se and/or Cu-Se and Au-Se stretching vibrations . Tentatively, we can assign a strong peak at 235 cm'1 as a Se-Se stretching vibration and another strong peak at 217 cm'1 to a M-Se stretching vibration. The spectrum is shown in Figure 68. It is usually difficult to interpret the IR spectra of metal polychalcogenide compounds without ambiguity because the Cu-Se, Au-Se, and Se—Se stretching frequencies fall in the same low-frequency lR region (ZOO-340 cm”) and that systematic lR spectrosc0pic data for the various free ligands (0x2: x=2-6) and metal chalcogenide complexes are still lacking. 9f” 83“ with!!!" 1‘42““ alewarsowmmmw vgotortqtom still-elbow him lmm Hm;:_i 75535.3 ("TH") of erroitsgiteevm 19mm? 1“ . ._ ‘ I ’v’Sw Vebn'J 915 sum (law '3 ..t . ; ."',‘f“..v"(-.. amortve .98qu no'tga't ”fl * ,. . .. l ;) Man 5868 or sub L 'f cf‘..““..i' " 311'. i332bn300'. _;r; v- ” l ‘ . ' is \Q QI‘O‘SB ‘Smm a.” , . _m-l...’ «. ..rt' . 3 1 ,~ ~~ add at {more el "M'fl -, l; ..3g.'-.;Esr‘:)vloq mom b" , l. . ‘ -, Q‘s-’2 5:998 b.1338“ “ u . . ... :5 1: ..., 3.); “'vli,‘ DMZ-00$) w. x . , ,P» _ _. Ft "-2.53; ebwsgil out a“ fi EULA” " 04‘ 31 ' angb:vv '1'!!an 293 5.. var 69.0.2 .0 888on min“. .8 2:9“. ...—8. cum—2:533 New Gnu «mu _ p _ on“ can l «.0» v. we (as) aaukusum § ! ¥.‘w _.....“ «...—...“.- 92'0 .- [Au(Se3)Se]n"' eq 1. However, the reduced chalcogenide ligands (Sez‘ in the [Au(Se3)Se]n ' chain) act as bridging ligands to Au3+ ions. These anionic [Au(Se3)Se]n"' chains are connected to each other by Cu atoms to form a stepped layer as shown in Figure 71. The Cu atoms are .9? 3817311. i: oéensmlb- 991.11 befeoilqmoo 1m """OT-f: ‘15? SW . vilsufq "901100 .MWMM ' _ T .J? flair;- *1 I) Deflflff 259.110 "datum . 7' “Q'- * ’.l( 1 n: nvrni‘a sis nlsrto - - . «i. r L"' .: 31 '2 story?! wen! siv an“ M ~ r qr . - ‘ new, .3 918 among UA lo gum-u x . ' 1:3": nu. ,iwvnvaottneonlh ., 1. " ..m. moi; r v) $918001 51: M“ . , 1:.iw "l swans oi m l, 9 ‘1 { ~‘l‘tea'7'1c/3-L. .‘ .218» 3‘ {amstw 2,. . - ~ " . -- r \ \,‘\8¢.3 1.06393011115- V.- "f; ; (a! a. L.) vwoni or" 01 w J _f 91.34,]. .».~ ' :3 77113;] "'€niq(aoe)(m ..‘I; . ."hees avast-1x0 118 as w ..l . .y 1 '.;;.-,r1 7.5-E; ‘ m1?) ‘"n{(g-98)UAIM l t: - ....— 'a9'€!‘;\n + '"rfl'cseWAl ( (men 1 " .3c._rc~3}u/ } : "if :7! S88} 21735935 9 HegwIBFio Oeoubet on! .W (41105 *é'UA 0! abnspil m* u; '{d tsr‘fo .1959 of befsennoo 918 anusrls "M1 .. (sew) olnoinu earl? ‘ rl .fT so 71 m nwon: 18 a ' ens emote to e T 1 or e as ye oeqqola I mol 0”“? 91;... ; . . , . . . . . .55.”...33 . . . V 1 In) 1 . ‘ l . .. . .12.... ...._,..,.,.. 2.1. ”.5 111 I 296 Figure 70. Two views of the [Au(Se3)(Se)]n"' chains with labeling scheme. 297 Figure 71. ORTEP representation and labeling scheme of one layer fragment of AuCuSe4. ' i .a . ‘ 3 E v‘ .‘- i F -... ' ‘1‘. . i Z "M‘ . O V. Q ails:- gnlledsl‘or‘e newsman-am ‘iSTRO .tYM - , taeuom to M "f_. 298 bonded to two terminal selenium atoms (Se(1)) of the 8e32' ligands of one [Au(Se3)Se]n"‘ chain and to one internal selenium atom (Se(2)) of the 8e32' ligands of a neighboring [Au(Se3)Se]n"‘ chain, resulting in a neutral two- dimensional AuCuSe4 layer. As a result, Sea?“ ligands are bonded to five metal atoms simultaneously as shown in scheme 1. To the best of our knowledge, this is the first example of such ligation of 8e32' in the known polychalcogenide chemistry. Au Au 83(1) Se(1) Cu/ \ / \C Se(2) U Cu Scheme 1 In the AuCuSe4 layer, three-coordinate Cu atoms, situated on a crystallographic mirror plane, expand their coordination geometry to become tetrahedral by Cu- Se(3) bond formation between the layers, producing a three-dimensional structure. This creates two types of empty tunnels which are composed of 10- membered rings and 8-membered rings. These tunnels are quite small; the short dimensions are 3.254(3) A for Cu--Se(1) and 3.283(3) A for Se(1)--Se(1) in the 10-membered ring, and 4.036(1) A for Se(3)-Se(3) in the 8-membered ‘ ’ —‘ .F'Iwr 11 owt lsfluen 3 hi ohm-1 m M evil of bebnod sis abnopil 4598 M. d. 1;; tr: 'aasct em 0T 1 emedoe m am so - 1. iv. rt: 1:191! Have to slam 1813' “"3 ; .1. --’ n -.;s?: .yttaimsrto . ‘l I‘ . I. 9 z '11 ~ .' ' 103 ' 5 f \ Q“ ‘ ~ . US) 9 l r. «.2131; .‘.:' 1 _ r '19.: t 1111311598113quan gt: .3 lob-.4.. 0711 c: {1"‘11Fli‘x n ‘9: 111710170 'liSfii onnqxs ,anslq mini ".ezowterl nonsmtot bnod (8)08 it xe: 11311: .5 rgmoubiwq 3‘78‘.':‘ 0: to '39.; 13.11110; we 11.3an 219nm \. 'k,‘ 51.: earn? c-w: 3915913 elrlT ..m an: ,lusmra 512.}; uni. venom 983:" .1941: ream-ewe bns egnl‘l W (flag-(l 158 10.1 A (8)8898 bns (use :1 31.11 l. l.» 12838 315 enolznuhlbm betodmoma en: n1 (8)92--(8)98 to! A (rt-9.80 ls bns .gnn betedmom-Of "I“. . 2 < ' - ‘- 299 ring. It should be noted that one of the tetrahedral Cu bond angles (Se(2)-Cu- Se(3)) is quite large at 152.3(1) deg. This is probably because the Cu atoms are derived from a linear rather than tetrahedral coordination. The Cu-Se(2) and Cu-Se(3) bond distances are relatively short at 2.423(3) A, while Cu-Se(1) bond distances are long at 2.528(2) A. Selected bond distances and angles are given in Table 90. 300 Table 90. Selected Bond Distances (A) and Angles (deg) in AuCuSe4 with Standard Deviations in Parentheses Au -Se(1) Au-Se(3) Se(1 )-Cu Se(2)-Cu Se(3)-Cu Se(1)-Se(2) 2.475(1) (x2) 2.483(1) (x2) 2.528(2) (x2) 2.411(3) 2.423(3) 2.396(2) Se(1)-Au-Se(1) Se(1)-Au-Se(3) Se(1)-Au-Se(3) Se(1)-Cu-Se(1) Se(1)-Cu-Se(2) Se(1)-Cu-Se(3) Se(2)-Cu-Se(3) Au-Se(1)-Se(2) Au-Se(1)-Cu Se(2)-Se(1)-Cu Se(1)-Se(2)-Se(1) Se(1)-Se(2)-Cu Se(1)-Se(2)-Cu Au-Se(3)-Au Au-Se(3)-Cu 180.00 (x2) 9826(5) (x2) 81 .74(5) (x2) .— 95.5(1) 101.40(8) (x2) ‘ ; 97.12(8) (x2) 1 152.3(1) L 104.28(6) 100.050) 98.570) 101 23(9) 105.190) 105.190) 97.13(6) 114.44(6) (x2) (l)a€~uA-(t)’08 (8)98-uA-(Mee [5' '19:), lJ,¢\-' l ‘92 ". l / . 5" l a . ‘2 1 - I (anus (8)8958 301 Based on the oxidation states of the metal atoms (Cu1+ and Au3+) which give rise to a completely filled valence band. AuCuSe4 is expected to have semiconducting behavior. However, no effort to investigate the charge transport property of this material has been made due to small size of single crystals. Preliminary attempts to measure its band gap in the mid- and far-lFl range did not show any spectral absorption. Work to grow larger single crystals for charge transport property measurements is in progress. One interesting feature of this material is the charge neutrality of the framework; it does not require charge balancing counter cations. The fragment of this structutre, the one-dimensional [Au(Se3)Se]n"' chain, would require charge balancing counter cations to be crystallized as a one-dimensional chain. However, the incorporated Cu atoms balance the charge, form covalent bonds with the chalcogenide ligands of the [Au(Se3)Se]n"' chain, and yield new structural framework. Even though the structural stoichiometry of AuCuSe4 is same as that of the known AuCuTe4, AuCuSe4 has unique structure. The structure of AuCuTe4117, which has not been characterized well yet, is known as the derivative of AuTez118. This unique Se32' containing compound of AuCuSe4 underscores the great potential of the mixed metal system to explore new chalcogenide materials with unusual structures and perhaps interesting properties. ltoqanatf» omens DEW Cm.) alstev‘lo sigma to 9le M51301 I.“ m I“ - ; 3ng5! Hl-‘SY ans -bt'l:t erlt m an mo ail om ‘ .~ 1,1; 1,. 1131):) 93pm: 1519153 Hill; 0‘ )lWOW "0M 9,913.1.»4} q '51 afnemom .n l. ' ', - helmet”: a.“ 11 911186“ 911mm!“ "A “A. l 1'." . l , . ... muskeut‘pellonasobl: . 1’ '4. Jr’nLA: s l1.|l¢’"v."’l' 3; r; 9111 31103001.“) 1 ,- . . 7, .- - ' f ' 1) ..-lnuo: pumice “3 1 4 . '. 3'1‘t' 1l .tt'-1(:(ltooniom.M ,3)», ,.,l l .391 . scmsgoalsflo Of. . . . 1 1- n1. . o.“ iwowemsfi m -. 1,, ,_ ~- , ‘ [nix-’51flolsmalm .. » l, .1 4 ~ .. "l'..eTll£)uAloflM 4 - ,_ _ e : . .l"‘ to tavlmvltobfl" .,._ 3:8 , . . Mozaemu m » -. . 2, ~ railwasr-l ablneooolubm .m CHAPTER 9 Concluslon In this work we have demonstrated that a new class of chalcogenide materials is accessible at intermediate temperatures using low-melting alkali- metal polychalcogenide fluxes as solvents and reagents. The alkali-metal polychalcogenide fluxes are now proven to be a significant synthesis tool in the field of chalcogenide chemistry. The low-dimensional polychalcogenide compounds presented here are beautiful examples of the structural diversity and bonding flexibility expressed by the polychalcogenide ligands in the polymeric solid lattices and rival that found in soluble molecular species. Especially, the redox chemistry associated with Au1+’3+ ions in the polyselenide (Azsex) fluxes represents the complex equilibria existing between various polychalcogenide species in the fluxes. The incorporation of structural units normally encountered in molecular chalcogenide complexes into polymeric structures should stimulate the interests of both molecular and solid state chemists. This, essentially solution, synthesis technique may eventually allow reaction control, currently available (only to some extent) in molecular chalcogenide chemistry. It could serve as an interface area between high- and low-temperature synthesis and should help bridge the ever narrowing chasm between molecular and solid state chemistry. Low-dimensional monoselenide compounds and intriguing telluride compounds with two-dimensional framework to complicated three-dimensional 302 21. ms? 04331: norm :3 overt ow tho. “ -l'.'. - L55 2 'Ju W alpacs’mwm ’8 sum . , w i ,; .1 rs- ‘21 we» :a ’99 29mm ebmm ~ 1 m' r. 1 l» 5 i1" ' 81L» ‘cEY'fn 0 ' ' l~ 'l '2 ; 'lH -- if Gabi-169003019 .yy 1.. ,. -,. 7 ' it _‘h’K' 91F "‘ ‘3‘?" ‘989‘q- 1 ,, I. ‘l n 31297? 'C‘ 55131 X3" WW“ .4.. , r- a t; 1v 2 .1: ‘ aoit ei biloc , ,4 r . .1 ., “mt, w S. rain.) Afr/31 em, N ,‘r‘ . , 4. w w ‘ Ji‘lSi'QngST eoxufl ( all! g. , r=i4vflfvll18 ebine ‘ r» ;.;r:=,.l.;t l. .2 : " ' ‘ _ ”w. 19313an m 3133:- billfc :5: mmm.) :- _- I Murine bluofla 00m wol'r; yileufnem vs”! . vwzmeeee .aiflT .m ‘Slui-alom m ,iwawe 4:“. , 1.1; w -I-v l‘lrwe'rzus ,lonnoo am one 49m memes.- :91»; re “-1 e.- w. Lama :' mam m mastic Qf‘leO’fiBfl fave .-,.«~.:- . w. -~ blur-rib His gleenfnva MW 33 man: 3.5.2; u :03 one teluoolom ”t . ebimilet pniugmm nus ahnuoanr elmlawman-m lsnoiammib-w bnolenomib-ee‘lrirNavigator-1c; o: howsmmi A ()5 .r it1rrruanamib-OW! m. 303 cluster framework presented here show the remarkable structural complexity associated with multiple-bonding characteristics of chalcogens. The interesting electronic properties from semiconductors to metallic conductors highlighted the validity of this work in exploration of new chalcogenide materials. This intermediate temperature regime could serve as a new synthetic frontier where one can stabilize totally new structural chalcogenide compounds with perhaps unusual properties. These materials are considered as metastable or kinetically stable compared to the thermodynamically more stable high temperature counterparts. The successful expansion of this method into mixed chalcogenide systems and mixed metal systems has shown another potential approach for the synthesis of new structural chalcogenide materials by incorporating two different sized chalcogen or metal atoms with their own preferential coordination environments into the polymeric lattice. The remarkable ability of (poly)chalcogenide ligands, to form multidentate bonds to metal atoms and to catenate, bridges different substructures in two and three dimensions and makes the very large number of compounds accessible by this synthetic route. It is evident that this is an exciting new synthetic area and thus far only the surface has been scratched. The bulk of this work and even more interesting chalcogenides are yet to come. . 5 am borngilrtgfd em MW Pin" alshetsm sbinegoalsflo m b m ‘ was My :elmo‘l oifenlnve wen s as emu M m -- ~‘1vw .mhngmjr'. :rU-lemolsrto letufau'm m M .,‘, '7,““1"T77:'.l"".l at: clstvefsm sesrlT . ‘- -.r:~._...;r"1" ’2 W" of carammoo . . ' #18qu I ..- r“ . H ll i:;.',"¢.(:‘.m um.“ . N l . t] .1 , ~ '_ ”a 113V? H 1m boxim -- -- l 5. {H gr: 5“" "l. ' m '0 l . - t ' w .c D' olsno but. I“ .; )l' 7.2 “' ‘Eowmu‘Jflem , , 1 ,1" . -lQ:s.‘l‘left01 0017"“ , r,. , r it abrtod M ...: s - . ... '.rl" Ck».- l‘l' om cw? ni 89M M .: . »' readiazecosw ,. - -... 1. ». . mm rneblvo db’MV 25,: . r; it ""3? mad can 03“. moanogoolado an“ LIST OF REFERENCES (a) Rao. C. N. R.; Gopalllakrishnan, J. New Directions in Solid State Chemistry; Cambridge University Press: New York, 1986. (b) Hagenmuller, P. Preparative Methods in Solid State Chemistry, Academic Press: New York, 1972. (a) Corbett. J. D. In Solid State Chemistry: Techniques; Cheetham, A. K., Day, P., Eds; Oxford University Press: Oxford, 1987. (D) West, A.R. Solid State Chemistry and Its Applications; John Wiley and Sons: New York, 1984. (a) Massoth, F. E.; Muralidhar, G. In Proceedings of the Climax Fourth In ternational Conference on the Chemistry and Uses of Molybdenum; Barry, H. P., Mitchell, P. C. Eds; Climax Molybdenum Company: Ann Arbor, Mlchigan, 1982, p 343. (a) Coucouvanis, D.; Toupadakis, A.; Koo, S. -M.; Hadjikyriacou, A. Polyhedron 1989, 8, 1705-1716. (b) Clegg, W.; Christou, G.; Garner, 0.; Sheldrick, G. M. Inorg. Chem. 1981, 20, 1562-1566. (c) Pan, H. W.; Leonowicz, M. E.; Stiefel, E. l. Inorg. Chem. 1983, ,2_2, 672-678. (a) Draganjac, M.; Simhon, E.; Kanatzidis, M. G.; Baenziger, N. C.; Coucouvanis, D. Inorg. Chem. 1982, 21, 3321-3332. (b) Coucouvanis, D.; Swenson, D.; Stremple, D.; Baenziger, N. C. J. Am. Chem. Soc. 1979, 1_Q_1_, 3392-3394. (C) Miller, A.; Krickemeyer, E.; Reinsch, U. Z. Anorg. Allg. Chem. 1980, 4_7_Q, 35-38. Kanatzidis, M. G. Comments Inorg. Chem. 1990, 1_0_, 161-195. (a) Ansari, M. A.; Ibers, J. A.Coord. Chem. Rev. 1990, 1%, 223-266. (D) Kolis, J. W. Coord. Chem. Rev. 1990, 1_05_, 195-219. (a) Steigerwald, M. L.; Rice, C. E. J. Am. Chem. Soc. 1988, 119, 4228- 4231. (b) Steigerwald, M. L.; Sprinkle, C. R. Organometaliics 1988, 1, 245-246. (c) Steigerwald, M. L.; Sprinkle, C. R. J. Am. Chem. Soc. 1987, 109, 7200-7201. (d) Chemical Perspective in Microelectronic Materials; Gross, M. E., Jasinski, J. M., Yates, J. T., Eds; Materials Research Society Symposia Proceedings, 1989, 131(references therein) Eichhorn, B. W.; Haushalter, R. C. Inorg. Chem. 1990, 29;, 728-731. 304 3638i :I::.'~ Ins-:4 l . ‘. n. /- 13.-'5‘“ gvfliflfiqofl .q . '5 y‘ - I~I.,.:.’j. g-‘~‘I\.I.Jem(_‘ we») ... 080"!th 9" .9..- :2 IItlsaIeme egbndmso. w_ 5459.?) INC-:- . » -' . _. pr novwolrt M. .. _, ZIP rt , 7.. —,"' ., 3...,1343 “,P'HI 9‘03 ”I G ..L ' _ ‘.‘"”-’ “ :;". . ' " I‘ 'V1'.-3l|:‘-"' J JYJVO I863 U“ ,. _I. ‘. ~ '. .. 3;. has III-atmedi) 9186” I,” .mr- ,‘ I‘I‘ ‘ "I, t r; "... I}: ."II I U I \ é) ‘ H“ 5 l a “F-O:) \Bf‘d‘ r ' :1 11““! q H 7 " J I i --.'.?".‘:'I nsginolM Jodi.“ I ‘ H-.,.;I 'J zirrxxvuoouoo 0* . ~_ *"I ’Jsernmbe 3W _. . . .. III a neonatal! ‘~“~ ‘ 0' n. ..I1I ‘ I ‘: U,- ‘ M .W IMW - . I “ . III-.0 ; .\ cameo-10 (no; I . _ ‘ it") .r-WI n,“ m n. Bin 8 , ..;I*:‘ noanow I1: I , 7. J t - ‘. ... brat-5242988 .LQI. «WW I ”-1 III’I‘TI‘Y r0903 .QNK .M , -;. a . .-Ir:~{'.,r:1.I::) i) .M ,8ibw_, a J‘s" r~ . .5'Iz' 75: ..., we? III“ v 1‘ " ”J“ A M kW 3' 3’ GHQ-5;,— pf; 09%;; :9” I‘i“~f"-:‘ NooO W L Hwy... '3 .z‘. ,r;.I .~ sear 3:. . .. .. a . ; M blemeoiotefp) \ 8887f?:er_rI.I.s‘Ic-.m:u.;3t--I"; .) ~Ii"'IID_?T' . - 5‘ I'IILEW‘SQB’aml I’M“ "'J . ‘ . ......t . -l 1w "t $4678 (0) 3‘3“ “43;; V89? 3;;i met?» HA I 5-1 ._ «mi. 5;; ,I 3)an .9») watt: era- IIIII. Isuzu-I3 Is; It???) (b) lOSV-OOST 091:0? riotseeeFI :lsnmsw . tit. *9): Y ..le J ifianiasb .3 M -. ,‘ ‘.‘.'-8. mil 28C"? 1‘. 1);." . 989i 39”.qu m ”g? - 15.397 g3 peer 11:90 I L» oI‘. 3 R ,Ietlsrlausl-l .W 8 ,M' ., "J'l 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21 . 22. 23. 24. 25. 26. 305 You, J. -F.; Snider, B. S.; Holm, R. H. J. Am. Chem. Soc. 1988, 110, 6589-6591. Marbach, G.; Strahle, J. Angew. Chem. , Int. Ed. Engl. 1984, g; 246- 247. Wardle, R. W. M.; Bhaduri, S.; Chau, C. N.; Ibers, J. A. Inorg. Chem. 1988,21, 1747-1755. Herrmann, W. A.; Rohrmann, J.; Herdtweck, E.; Hecht, C.; Ziegler, M. L.; Serhadli, O. J. Organomet. Chem. 1986, 81_4, 295-305. Coh, L. Y.; Wei, C.; Sinn, E. J. Chem. Soc., Chem. Commun. 1985, 462- 464. Rueingold, A. L. ; Bolinger, C.M.; Rauchfuss, T. B. Acta Crystallogr, 1986, 429, 1878-1880. Giolando, D. M.; Papavassiliou, M.; Pickardt, J.; Rauchfuss, T. B. Inorg. Chem. 1988, fl. 2596-2600. Eichhorn, B. W.; Haushalter, R. 0.; Cotton, F. A.; Wilson, B. Inorg. Chem. 1988.21. 4084-4085. Haushalter, Ft. C. Angew. Chem. Int. Ed. Engl. 1985, _24, 433-435. Fritz, P. M.; Beck, W.; Nagel, U.; Polbom, K.; Herrmann, w. A.; Hecht,. C.; Rohmann, J. Z. Naturforsch. 1988, £3_B, 665-670. Scheidsteger, 0.; Huttner, G.; Dehnicke, K.; Pebler, J. Angew. Chem. Int. Ed. Engl. 1985. .2_4. 428. Kanatzidis, M. (3.; Huang, S. -P. Inorg. Chem. 1991, 30, 3572-3575. Kanatzidis, M. G.; Dhingra, S. Inorg. Chem. 1989, g8, 2024-2026. Kanatzidis, M. G.; Huang, 8. -P. Inorg. Chem. 1989, 28, 4667-4669. Strasdeit, H.; Krebs, 8.; Henkel, G. Inorg. Chim. Acta 1984, 89, L11- L13. Dhingra, 8.; Kanatzidis, M. G. Unpublished result. (a) Kanatzidis, M. G.; Huang, 8. -P. Angew. Chem. Int. Ed. Engl. 1989, _2_8, 1513-1514. (b) Huang, 8. -P.; Kanatzidis, M. G. Inorg. Chem. 1991, 1), 1455-1466. mwmwfiw mgsmngamm. a!!!” may"; :g-w .A L and: ,M .0 .mo ..o .mgd‘fl . ,3 .= ‘1 Lucknl , ma 0‘” .~ ’ U ,irtoah Tilt-89.: 15:8. 889’ _ L44 rlfiz ------- 7“ , i, 2' I ‘54): l‘ 'V | x i‘ .r..,r J ‘ { “)3 “, ‘0‘. 9. O. .- 'X ." .A ' ..\ ,_, ' ‘34 }. :3 ‘ 5. / ~::"" "2.9": '«‘ x P i ‘ 4?, v' ‘ ->,¢n.qs« M O M “PM“ \_2 mun-Q ’fl. 0.. 3,»me L W A” want) 19W.» ("3'10 3 macaw; . zagnflo‘! . .J .A . « 8' «30: A839 . ’L-U 9 ’12?“ 5*; b .‘a OBEY-8 .5. ‘ ...V‘C .8 'J .fi new ,4 quash! my .nm .4 \_' 4.‘1999W ",3 939er L‘..‘ M W ‘ 3 a ~13!wa eMeibimm «11 1d) flab-")4 ;.H 3M , .MJ 3 moans)! , 8 m ‘ ., e M 215mm: warmer .35”th q. iiifii ‘ v 75 J 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 306 (a) Alonso Vante, N.; Jaegermann, W.; Tributsch, H.; anle, W.; Yvon, K. J. Am. Chem. Soc. 1987, 111, 3251-3257. (b) Weisser, 0.; Landa, S. Sulfide Catalysis: Their Properties and Applications; Pergammon: London, 1973. (c) Chianelli, R. R.; Pecoraro, T. A.; Halbert, T. R.; Pan, W. -H.; Stiefel, E. I. J. Catal. 1984, 8_6, 226-230. ((1) Pecoraro, T. A.; Chianelli, R. R. J. Catal. 1981. 62,, 430-445. (a) Smith, R. A. In Semiconductors ; Cambridge Unversity Press: 1978, pp 438. (b) Barett, B. E. et al. Infrared Phys. 1969, 9, 35. (a) Mickelsen, R. A.; Chen, W. S. In Tenary and Multinary Compounds: Proceedings of the 7th Conference; Deb, S.K., Zunger, A., Eds; Materials Research Society, 1987, pp 39-47. (b) Steward, J. M.; Chen, W. S.; Devancy, W. E.; Mickelsen, R. A. In Tenary and Multinary Compounds: Proceedings of the 7th Conference; Deb, S. K., Zunger, A., Eds; Materials Research Society, 1987, pp 59-64. (a) Whittingham, M.S, Science1976, 192, 1126-1127. (b) Whittingham, Ms. J. Solid State Chem. 1979, 29, 303-310. Bellman, A. A.; Byer, R. L.; Eimerl, D.; Feigelson, R. S.; Feldman, B. J.; Goldberg, L. S.; Menyuk, N.; Tang, C.L. Applied Optics 1987, g, 224- 227. Rosenberger, F. In Preparation and Characterization of Materials; Honig, J. M., Rao, C. N. R., Eds; Academic Press: New York, 1972. (a) Elwell, D.; Scheel, H. J. Crystal Growth from High-Temperature Solutions ; Academic Press: London, 1975. (b) Molten Salts Handbook; Janz, G. J., Ed.; Academic Press: New York, 1967. (c) Mamantov, G. Molten Salts; Marcel Dekker: New York, 1969. (a) WIlke, K. -T.; Schultze, D.; prter, K. J. Cryst. Growth 1967, 1, 41-44. (b) Larson, G. H, Sleight, A. W. Phys. Lett. 1968, _2__8A, 203- 204. (c) Sleight, A. W.; Jarret, H. S. J. Phys. Chem. Solids 1969, 29. 868-870. (d) Harsy, M. Mat. Res. Bull. 1968, 3, 483- 487. (e) Shick, L. K.; van Neida, A. R. J. Cryst. Growth 1969, _5_, 313- 314. Heraldsen, H.; Kjekshus, A.: Rost, E.; Steffensen, A. Acta Chem. Scand. 1963, 11, 1283-1292. (a) Burschka, C. Z. Naturforsch 1980, 358, 1511-1513. (b) Gattow, G.; Rosenberg, 0. Z. Anorg. Allg. Chem. 1964. 332, 269-280. Haushalter, RC. Angew. Chem. Int. Ed. Engl. 1985, gt, 432-433. Banda, R. M. H.; Craig, D. 0.; Dance, I. G.; Scudder, M. L. Polyhedron 1989, 1g 2379-2383. are: 23914 911219an 998mm :W 38 3 .889! 2.1649me 3 I i-"~‘ - ’J'i'v)"; ‘I‘IZ‘II’II’AUM "0M“, J‘s \mT hi 8 W (9930 ,5 fl ' ‘19?” . "1 1“7.3~"IU: . H. '9” SCREWS 111‘ “Uh ‘ '1; _ 1'12"“57 . ‘ 1,. l“'_‘~‘.‘I."jY:'~ Ll‘ Y'P'fft, 3? I397 , I. - " I' ‘- W" .-..,-, r" ’-— '1 meleme .3 .W, ‘ 7 ' ,3"? ..i-\'.=.'I‘~.I:.‘-"; 'M‘. 3m 19 3. . ' ‘ ' “3| Wet-w". (13168209 l ~-,,. ' :C L"; ' EDGE“. £- :1'58 Ml! 5 - .1 ', . r' . ‘7 bl L .. l . ..- ‘. 1: ‘~l-v _, i ”we _ z.‘ A In“ ‘, ..-. , ~ ,.I :1, 9,1,3 J 9M m >,- “:qg‘rq' rt 4 , 1&th .M'? ,I I‘ " asR..M.l. v. . .~ I 49 Is? . ".3 "st (3) - - . . > . ,_ y“ z-ir. Ibis-M ' 330111308 ' I , ""'"I't~.vt.:"- 1,9,“. " . . ...1 k; - 1i." ’IIw-I: WEB 1‘“ -‘f -‘ ,. If": ’7 71.9““) e ,' , . ‘ ‘0 ‘1 i: .71me (d ' " ' ‘ , J 1:31.; . YU'V A j :1 an»? . ,I‘.J ("F‘ Y‘EVVI 5J1 YM (b 5. Ice? rinse: .-',. ’._‘I .\ 3.. an“: I}: I. . _r': ‘. ‘. ‘_r-‘.."'I.~l3_?:", 1 SEC-ru’ Ex do)!“ ‘ 1’ New 883. 31 not .'-‘_ ",""_-:‘.\T' "(41 7": El; ” E 1‘. _fi’f a tip! 141:, , TAX; flinfi S 3 3“,xw( ‘ case»: 13;; mm mw . .I..-..IA 0,91 SELL-TIE} 3.; see; '._.II.:I 3‘3 Im mmm wegnA JR m i ,. , _ A {Wantnfloq ._l 1.1 1.9531133 I; I ,sonab :C @1813 :H .M .R M l v ' 's'. Ii, 7» A_-_ __"_, - 1‘ - .. ' 'h‘ I: ' 1‘ III—I: at 28999189 31,." ,- 39. 40. 41. 42. .3 46. 47. 48. 49. 50. 51. 52. 307 Kanatzidis, M. G.; Huang, S. -P. J. Am. Chem. Soc. 1989, 1_1_1_, 760-761. Haradem, P. S.; Cronin, J. L.; Krause, R. A.; Katz, L. Inorg. Chim. Acta 1977, 25, 173-179. (a) Brec, R.; Evain, M.: Grenouilleau, P.; Rouxel, J. Rev. Chim. Miner. 1983, 29, 283-294. (b) Brec, R.; Ouvard, G.; Evain, M.; Grenouillear, P.; Rouxel, J. J. Solid State Chem. 1983, fl, 174-184. Proceedings of the Fourth International Symposium on Molten Salts; Blander, M., Newman, D. S., Mamantov, G., Saboungi, M. -L., Johnson, K., Eds; Electrochemical Society: Pennington, NJ, 1990; Vol. 84-2. Diehl, R.; Nitsche, R. J. Cryst. Growth 1973, 29, 38-46. Scheel, H. J. J. Cryst. Growth 1974, 24/25, 669-673. (a) Sanjines, R.; Berger, H.; Levy, F. Mat. Res. Bull. 1988, E. 549-553. (b) Garner, R. W.; White, W. B. J. Cryst. Growth 1970, _7_, 343-347. (a) The Sodium-Sulfur Battery, Sudworth, J. L., Tilley, A. R., Eds; Chapman & Hall: London; New York, 1985. (b) Fischer, W. Mat. Res. Soc. Syrnp. Proc. 1989, 135, 541-551. (c) Powers, R. W.; Karas, B. R. J. Electrochem. Soc. 1989, 1_3_§, 2787-2793. Gmelin's Handbuch der Anorgnischen Chemie; Verlag Chemie: Weiheim/Brgstr., FRG, 1966; Sodium, Suppl. Part 3. Pp 1202-1205 and reference therein. Pearson, T. G.; Robinson, P. L. J. Chem. Soc. 1931, fl 1304-1314. (a) Mathewson, G. H. J. Am. Chem. Soc. 1907, 29, 867-880. (b) Klemm, W.; Sodomann, H.; Langmesser, P. Z. Anorg. Allg. Chem. 1939, 24_1, 281-304. Kanatzidis, M. G. Chem. Mater. 1990, _2_, 353-363. (a) Sunshine. S. A.; Kang, D.; Ibers, J. A. J. Am. Chem. Soc. 1987, mg, 6202-6204. (b) Sunshine, S. A.; Kang, D.; Ibers, J. A. Mater. Res. Soc. Symp. Proc. 1987, 91, 391-396. Kang, D.; Ibers, J. A. Inorg. Chem. 1988, $549551. (a) Kanatzidis, M. 6.; Park, Y. J. Am. Chem. Soc. 1989. 1_1_1.. 3767-3769. (b) Kanatzidis, M. G.; Park, Y. Chem. Mater. 1990, 2, 99-101. (c) Park, Y.; Kanatzidis, M. G. Angew. Chem. Int. Ed. Engl. 1990, Q. 914-915. .19r1'1M midi) M L EMW“ .. :11 15911113011918; .M 111016.91 +" 1%; 'A. , ”is v , w 5."?- ‘1 Ic' 1 ,. IcI mmaoqmy'a \snoIstm M 91‘ h \ . 11293.1" 1‘. L1 ,Ipnuodse .0 _VOMBmBM. a O . ‘ "E . ‘ ' 009! 1.51/1 wrigninnse 11191309. ' “'7‘” :19! 11.11116 hams .L 51.91139“!le .A ‘ 9‘s - . M"; mum?) W.) L L“. ‘-. , )q‘. I'N'fi ’, MN :J H P: «"8 .RR .1‘ r '15.: J '5‘“) ‘I P. x .911..-!';A . . ,_;I . . ' 33112.8 1111111?) .1199»? 09" . Q 5' 'I \ ii ‘5‘. ' NEH"! IJ37’ 1 ”8H 8 n . -. . , an 3‘4: ..‘LL 9274 .Icv' . acme .afl -.-... 1'5: 7. JL' saw 1;? 115.135nt \ -c"1e‘. .- :4.) 411‘J‘1'U‘“ 8mm .“ ‘ ~I., .— ,.,.;- we? a" 1 nemamlorlbw ruined! eon-10M g} , -. ' I." \,I_,IVJ~ “-91 ', 3‘2”” a) o. _ 5,, - I: I'mgmoboe ;.W as: 106-{fl - '_ «if ”3311?“ 3911!”. r‘ M am .. ‘- 4,31 .‘;(f :1» ;I;II ‘ - '_. 3‘31? {2179' " 8 .emrianuefl) '" , ~ . - , ._ 1:. cwéi (a; .mm ... M1 1‘35: 189! .3019 a” 9.3-€96: 1 .8351? men; .;--I-; .1 A I. ”and! .0 ..fl 1 . 91511;) mi- ; Y 9:139 ; (3 M mun-1018 ; 1. 1 J I“ x J ,I £9995 W.099l.1191e-A mad"; Y 11189 B M um 23 MEI-1'3 133111‘.m'311" w'egnA E) M W 55. 56. 57. 58. 59. 60. 61. 62. 65. 308 (a) Park, Y.; Kanatzidis, M. G. Chem. Mater. 1991, 3, 781-783. (D) Park, Y.; Degroot, D. C.; Schindler, J.; Kannewurf, C. FL; Kanatzidis, M. G. Angew. Chem. Int. Ed. Engl. 1991, fl, 1325-1328. (a) Meerschaut, A.; Rouxel, J. In Crystal Chemsitry and Properties of Materials with Quasi One-Dimensional Structures; Rouxel, .J., Ed.; D. Reidel Publishing Co.: 1986, pp 205-279. (b)Superconducticity in Ternary Compounds; Fischer, 0., Maple, M. B., Eds., Springer: Berlin, 1982; Vols. I and II. (c) Chevrel, F1. In Superconductor Materials Science: Metalurgy, Fabrication and Applications; Foner, S., Schwart, B. B., Eds.; Plenum Press: New York, 1981; Chapter 10. (d) Whittingham, M.S. Prog. Solid State Chem. 1978, 1_2_, 41 -99. Smith, D. K.; Nichols, M. C.; Zolensky, M. E. POWD10: A Fortran IV Program for Calculating X-Ray Powder Diffraction Pattern, version 10. Pennsylvania State University, 1983. Walker, N.; Stuart, D. Acta Crystallogr. 1983, 39_A, 158-166. Sheldrick, G. M. In Crystallographic Computing 3; Sheldrick GM, Kruger, C., Doddard, 8., Eds.; Oxford University Press: 1985, pp175-189. Frenz, B. A. The Enraf-Nonius CAD4 SDP System. In Computing in Crystallography; Delft University Press: Delft , Holland, 1978, pp 64-71. TEXRAY Structure Analysis Package, Molecular Structure Cooperation, 1 985. (a) Rfldorlf, W.; Schwarz, H. G.; Walter, M. Z. Anorg. Allg. Chem. 1952, 289, 141-152. (D) Brown, D. 8.; Zubieta, J. A.; Vella, P. A.; Wrobleski, J. T.; Watt, T.; Hatfield, w. E.; Day, P. Inorg. Chem. 1980, 19. 1945-1950. (c) Klepp, K. 0.; Boller, H.; Vbllenkle, H. Monatsh. Chem. 1980, m, 727-733. (d) Burschka, C. Z. Anorg. Allg. Chem. 1980, 4%, 65-71. (a) Takéuchi, Y.; Kudoh, Y.; Sato, G. Z. Kristallogr. 1985, 123, 119-128. (D) Fjellrag, H.; Gronvold, F.; Stolen, S. Z. Kristallogr. 1988, 1831. 111- 121. (c) Berry, L. G. Am. Mineral. 1954, 32, 504-509. (d) Evans, H.T.; Konnert, J. A. Am. Mineral. 1976, 51, 996-1000. (a) Burschka, C. Z. Naturforsch. 1979, 3% 396-397. (D) Peplinski, 2.; Brown, D. 8.; Watt, T.; Hatfield, W. E.; Day, P. Inorg. Chem. 1982, 21, 1752-1755. (c) Whangbo, M. H.; Canadell, E. Inorg. Chem. 1990 .29, 1395-1387. Muller, A.; Diemann, E. Adv. Inorg. Chem. 1987, 31, 89-122. Weller, F.; Adel, J.; Dehnicke, K. Z. Anorg. Allg. Chem. 1987, m, 125- 132. .. - m. \(‘flhfibflm . ) - rut-38 wegnnod 333 .5 .M "’3‘ q‘ amn‘nsm \MUbnoufiemZ (:1. fl 21'? ’3 E‘ .m-moe . 3 ant)? 21106 :-' " w: :3 . :r'urfif Eu) C-r wetnanO; ; 'b .52,“ MK) 7: -~ .4.-mm; ..o M W m .d— -*=“>wo'. tbs-J .Xf manna-33 to! 231%“ '33:?"1'H‘3U' $32 0‘ ‘ r gm :1 mmm. ..qu ”J ,hsu?a;M ,1“ "‘- ..\ “."-\u.,(vm .MOW 3... l‘)/ - \..,:J ‘_' 1153!.)03.m ..',« “1"3 {"3 r #13 “"3 ’3‘“: A 8 'M a; ' m4 . tin-C unuswgdwwo l : K. susxu-“l MM momma‘mfixa“ .& w. w ’ 14f- \ A w , . r" 3.3.-m» ,\/-.’ moms ) ‘ ..twu‘i ‘- ’,. mic?) fiat-t”. " ‘n .27 .... V" 3;.“6H T ”3W ..T ‘ 7 ' “flu? O N. ,(W (3 I.‘ 1:.v‘l'iffii'Mléj (hi .88" ; ' ‘ ' ~, ' ‘ ~.f,}'} '. 7,,- ‘ _...p ".M’ m w’ lubuéflfll 9 7:. ...“ ‘ "w *- ”_-‘ «1 )1; L\“4T‘.'3 D H gmtlofi _: ‘ H .N .3 g. - 7;; :2‘.’ +3?) wi.".:d"'.‘.5 "5- : \11198(O).f3! (““39 .3 3‘3? \‘2 9mm (0.“. A L flannel DU ‘ .z: :.:i« ”if: 622‘. g}; 3 2m fiaachutsm X I) .sfldoema (l) .- 3 saw :v >32? gm ‘7 3W3 ;3‘ W (7:92?th T ,ISBW . 8 .0 . A‘ 9;; 098? rm 4:3 mom .3 .tiebsnsO 1H M 0:19:1de (a) bat-QB?! met-ace: SS’QB J; war mad?) mom .vbAa 7,;nnam900 .A A“ . .es: 4»; tom mods gm .monA S )0 .exmndefl ;.L .bbA . =| .10!“ t .3! ‘ . 1 66. 67. 68. 69. 70. 71 . 72. 73. 74. 75. 76. 77. 78. 309 Nagata, K.; Tshibashi, K.; Miyamoto, Y. Jpn. J. Appl. Phys. 1980, 1_9_, 1569-1573. Krauter, (3.; Dehnicke, K.; Fenske, D. Chem. -Ztg .1990, fig, 7-9. (a) Jansen, M. Angew Chem. , Int. Ed. Engl. 1987, 2_6, 1098-1111. (b) Scherbaum, R; Huber, B.; Miller, C.; Schmitbaur, H. Angew. Chem. Int. Ed. Engl. 1988, fl, 1542-1544. (c) Kahn, M. N. l.; King, C.; Heinrich, D. D.; Fackler, J. P., Jr.; Porter, L.C. Inorg. Chem. 1988, _2_3, 2150-2154. (d) Fackler, J. P., Jr.; Porter, L. C. J. Am. Chem. Soc. 1986. 193, 2750-2751. (a) Dance, I. G. Polyhedron 1983, g, 1031-1043. (b) Hollander, F. J.; Coucouvanis, D. J. Am. Chem. Soc. 1977, 33, 6288-6279. (c) Chadha, H.; Kumar, H.; Tuck, D. G. J. Chem. Soc., Chem. Commun. 1986, 188- 189. (d) Coucouvanis, D. ; Murphy, C. N.; Kanodia, S. K. Inorg. Chem. 1980, 13, 2993-2988. Burschka, C.; Bronger, W. Z. Naturforsch 1977, 328, 11-14. (a) Burschka, C. Z. Naturforsch. 1979, 34_b, 675-677. (b) Haar, L. W.; Di Salvo, F. J.; Bair, H. E.; Fleming, Fl. M.; Waszczak, J. V.; Hatfield, W. E. Phy. Rev. 1987, 358(4), 1932-1938. (c) Fleming, Fl. M.; Haar, L. W.; Di Salvo, F. J. Phy. Rev. 1987, 338(13), 5388-5391. Savelsberg, C.; Schaler, H. Z. Naturforsch 1978, 33_B_, 711-713. Berger, H.; Eriksson, L. J. Less-Common Met. 1990, 1_61, 101-108. Savelsberg, G.; Schater, H. Z. Naturforsch 1978, 3L8, 370-373. (a) Muller-Buschbaum, H. Angew. Chem. Int. Ed. Engl.1989,_23 1472-1493. (b) Mflller-Buschbaum, H. Angew. Chem. Int. Ed. Engl. 1991, 33, 723-744. (c) Cava, Fl. J. Science 1990, 241, 656-662. (d) Rao, C. N.; Fiaveau, B. Acc. Chem. Res. 1989, 32, 106-113. Structural Inorganic Chemistry, Wells, A. F., Ed.; Oxford University Press: New York, 1987, p 538. (a) Whangbo, M. -H. In Crystal Chemistry and Properties of Materials with Quasi-One-Dimensional Structures; Rouxel, J. Ed.; D. Fleidel: 1986, pp 27-85. (D) Whangbo, M. -H. Acc. Chem. Fles. 1983, 1_6_, 95—101. (0) Canadell, E.; Whangbo, M. -H. Inorg. Chem. 1990, 2_9, 1398-1401. (d) Whangbo, M. -H. Electron Transfer in Biology and the Solid State: Inorganic compounds with Unusual Properties; Johnson, M. K., Ed.; American Chemical Society: Washington, DC, 1990, pp 269-286. Park, Y.; Zhang, X.; Kanatzidis, M. G. work in progress. may fino "fi‘ (d) .lll’l-SBBI‘ _ , m\ (“9613 10189851 - ‘ a mhnleH .0 9n“; .lM M ,- - -. 3 “‘I‘ . ."n Q-OEtS .83 .884" magi) gnth ’1 .' .. ... , Y, ad. 3'." so; east 6% menu m no Jada-fir»; . .. , AT‘ . . go" 53m t 31.-r .3 ‘,889t WW Dal :-‘..-_ ‘ We. ’13; wet 2.03 mm M .0, 133.324-. "(t-F «1880.80 0 mmm, "...-i T f ,qdowlul, C mnsvuomm i, I‘ . ‘ 3:. {'29 ”_.é‘fl’i‘I‘JISUI : n ,fig‘OK-‘U » LO .3“ d ‘ s VT- :2: NEW mustmd’»: "L 3 .sxm . J .113 45:35.) I} }_ I" ‘ I. H H “.88 Ila .q . I 2:15) 3. —.~'r V-twir cg; ,tsotyea nan ? ‘ F" “43.13.32 '75: w‘I .3‘ "3‘ L74 .011” r 1" 3.‘ ,. , a 07' 3:; J . until 3 9 .1.“ i ‘v ,,,_ .... Z. - 331513.?" .9) QM“ I “v 1 i; "UBCHSZUS‘WWu i} I . 4;; is ’L.2M Cd) SQ‘f-S‘ T . '3 9 ~ I") ”“395 Q53... 5- :r) 7f ,1:}\ ‘u‘II . ‘V.¥- ’3 {‘89ng ,.M .0 (M "I b; . 3 .- ... 3.3122232-«113 :mawonfiw $ (“id (1 YOU! .fl‘OY M 123.... ': f- .o aeshsmvfl. ‘mayxa‘ .3:2:s hut.) ‘ H- M 0W .W A at: £35.}? ‘-_.tm.-'«'i .C it)?! i. ,"8)-::'C.- ‘3‘. "330.3311? Isw-aiznsmfl-Bno- ‘- is: 2'3?- 5.9 51' .888” al.-EEK .rnsf‘xi‘ 33A H- M .0691de (a) .3.- (t. n).- -BR..' .33 099i magi) 910.4 H- M octpnsnw. 3, 91613 0808 am has \{gciaz‘a I‘ mm as“ nmim’i H M W . 93 .)l . noanrlou mamaqmq Isuzu: L‘ mw zbnuoqmoo m .aeseas'qq we: .00 nclgmrlasw 106in mm (and - 3891901: hi how G) .M ,zlblxtsnafl ;.X ,MLY‘ M . 79. 80. 81. 82. 85. 86. 87. 89. 90. 91. 92. 93. 310 (a) Roberts, L. E. J. MTP lntemational Review of Science, Solid State Chemsitry; Vol 10, p 241. (b) Bronger, W. Angew. Chem. Int. Ed. Engl. 1981, _2_Q, 52-62. (c) Meerschaut, A.; Gressier, P.; Guemas, L.; Rouxel, J. J. Sold State Chem. 1984, 31, 307-314. (d) BOttcher, P. Angew. Chem. Int. Ed. Engl. 1988,21, 759-772. (e) Krebs, B. Angew. Chem. Int. Ed. Engl. 1983, 33, 113-134. Klepp, K. 0.; Bronger, W. J. Less-Common Met 1987, _1_3_2_, 1173-179. Klepp, K. 0.; Bronger, W. J. Less-Common Met. 1987, _1_2_Z, 65-71. Klepp, K. 0.; Bronger, W. J. Less-Common Met. 1988, 131, 13-20. Rabenau, A.; Schulz, H. J. Less-Common Met. 1976, 13, 89-101. Structural Inorganic Chemistry, Wells, A. F., Ed.; Oxtord University Press: New York, 1987, p 414. Bronger, W.; GOnter, O. J. Less-Common Met. 1972, _21, 73-79. (a) Merz, K. M., Jr.; Hoffman, R. Inorg. Chem. 1988, fl, 2120-2127. (b) Jansen, M. Angew. Chem,, Int. Ed. Engl. 1984, 23, 1098-1109. (c) Mehrotra, R.; Hoffman, R. Inorg. Chem. 1978, fl, 2187-2189. Scheldrich, W. S.; Braunbeck, H. G. Z. Naturforsch. 1989, 44B, 1397- 1401. Sommer, H.; Hoppe, R. Z. Anorg. Allg. Chem. 1978. $3. 201-211. Toshikazu, O. J. Appl. Crystallogr. 1974.1, 396-397. Comprehensive Inorganic Chemistry, Bailar, J. C., Emeléus, H. J., Nyholur, R., Trotman-Dickenson, A. F., Eds; Pergamon Press: 1973; Vol. 3, p 311. (a) van Schnering, H. -G.; NOnIe, W. Chem. Rev. 1988, 33, 243-273. (D) van Schnering, H. -G. Angew. Chem. Int. Ed. Engl. 1981, 2_0, 33-51. (c) Schater, H. Ann. Rev. Mat. Sci. 1985, 1_3, 1-41. (d) Bronger, W.; Miller, P. J. Less-Common Met. 1984, L03 241-247. (e) Bronger, W. Angew. Chem. Int. Ed. Engl. 1981, 33, 52-62. Axtell, E.; Park, Y.; Kanatzidis, M. G. manuscript in preparation. (a) Donika, F. G.; Klosse, G.; Radautsan, S. I.; Semiletov, S. A.; Zhitar, V. F. Sov. Physi. Crystallogr. 1968, _1_2, 745-749. (D) Boorman, R. S.; Sutherland, J. K. J. Mat. Sci. 1969, 5(3), 658-671. (c) Barnett, D. E.; Boorman, R. S.; Sutherland, J. K. Phys. Status Solidi 1971, 4(1)A, K49- K52. .mmvr gin" WW'L... ‘ . «ea .33; net 19% mmmosmll .1. .w W: 3 “.uIi' ".7: 38M 19H: m mmoTP-m) L. W ,w; t ‘: ?e. 9.31”” WAN: L-v-J .. H .mmeJ - , ' 3.1.-3r: mm; a" . mmm“ M >12 \éfi‘" ,M ms:- r» 3.3-3332 ‘. C ,‘wl"bu w nuts. 9‘!» Rm) ‘3 . q “a 4-.I.!r-}! JI M A 31“.“ 0.“)! o ,i-l m w- I1 ~32.” .M .0“ f 3 1‘ 3 9:): R ‘Jflh)",g 111W 3 J. i 386 «"3 2 W dorm .lO-lf . ,7 ~ 2 :‘f’toH.H Jamal! l.'. \(a- I. 3' unrlirm‘l' wt... ’_ g...“ wwwafig‘lm :gfnlO‘i R 10w rte q.‘ . . . - ~ . - immune now (Q -_ -' ' - ‘r. ., x912"; ~. 43. an 1s.- fi'laa M gum '. 3. 3 ‘ ‘ :‘n‘ 3.6.9:? .3; Z.“ -' .3 nnA H mm 3 -- . . 3 . 2 ~. ...; 4:32“. :cmmo’.) 329...) L3 11:; 7. Itioi x3513 b?! It“ . 5.2“." vi;'.6-I-' .1”... ..:"3_.~T-. 3": "I (mn'Javisln: ‘r’ .X‘fiq ,3 m ‘ v swig .,: mi? ’5. 515.210.:233231 . 52.1791 . P, .r B rum palm} 95 £ng 31 o‘ 31.1;ch .7 (J1f90188 f-) I‘TB 8:13 13» 809: ”..k 15.1“ L )5 L .‘n 3.1-0.2 .i‘VeI twee amaze. amq A I bro: 8111113 ; 8 .R ,nuwg c 95. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 311 (a) Grey, I. E. J. Solid State Chem. 1974, u. 128-134. (D) Cohen, 8.; Rendon-DiazMiron, L. E.; Steintink, H. J. Solid State Chem. 1978, _2__5, 179-187. (c) Hoggins, J. T.; Steinfink, H. Acta Crystallogr. 1977, 13g, 673-678. Schils, H.; Bronger, W. Z Anorg. Allg. Chem. 1979, 156, 187-193. Bronger, W.; Schils, H. J. Less-common Met. 1982, &. 279-285. (a) Berger, H.; Van Bruggen, C. F. J. Less-Common Met. 1984, 99, 113- 123. (b) Berger, R. J. Less-Common Met. 1989, 14_'/_. 141-148. Berger, H.; Eriksson, L. J. Less-Common Met. 1990, 161, 165-173. Berger, R. Eur. J. Solid State Inorg. Chem. 1988, 25, 279-288. (a) Bernstein, J.; Hoflmann, R. Inorg. Chem. 1985, 2_4_, 4100-4108. (b) Hulliger, F. Structural Chemistry of Layer-Type Phases; Levy, F., Ed.; Reidel: Dordrecht, Netherlands, 1976. Klepp, K. 0. Z. Naturforsch. 1987, 426, 130-134. Savelsberg, G.; Schater, H. Mat. Res. Bull. 1981, 16, 1291-1297. Mulay, L. N. Theory and Applications of Molecular Paramagnetism; Boudreaux, E. A., Mulay, L. N., Eds; Wiley: New York, 1976, pp 494-495. (a) Anderko, K.; Schubert, K. Z. Metallkde.1954,4_5_, 371-378. (a) Patzak, I. Z. Metallkde. 1956, Q, 418-420. (c) Hulliger, F. Structural Chemistry of Layer-Type Phases; Levy, F., Ed.; D. Reidel Publishing Co: 1976; Vol 5, pp133-135. a) Jellinek, F. MTP international Review of Science Inorganic Chemistry Series one; Sharp, 0. W. A., Ed.; Butterworths: London, 1972; Vol 5, pp 339-396. (D) Folmer, J. C. W.; Jellinek, F. J. Less-Common Met. 1980, 78 153-162. (a) Hoffmann, H.; Zhang, C. J. Phys. Chem. 1985, 89, 4175-4181. (b) Burden, J. K. Chem. Rev. 1988, a, 3-30. (c) Tremel, W.; Hottmann, H.; Silvestre, J. J. Am. Chem. Soc. 1986, 198, 5174-5187. (a) R001, L. C.; Pennington, W. T.; Kolis, J. W. J. Am. Chem. Soc. 1990, fig, 8172-8174. (b) Adams, R. D.; Wolfe, T. A.; Eichhorn, B. W.; Haushalter, R. C. Polyhedron, 1989, _2_1, 701-703. (0) Haushalter, R. C. Angew. Chem. Int. Ed. Engl. 1985, 24, 433-435. .F Fifi 5 Hr. .. .... _, . .1 \‘ ....” ' - .75" .1!’ 7'.7 ,. 1. . ‘ j , ‘7 I‘ - f? ’ ‘ M 89MB? .682 .mnubfi.“ '38?- 979 £8 $89! .st nomM‘ .7 ‘ .1. .N ‘21.”? >39Ma157mxo’3-232‘ 1. 3. O human-V: .I. 9.2? ' r 77*. 380! 231.1 (12.111110368ij 1.8 .‘m .a'» ‘ ' (Lug? ‘.—‘_-’ "..".L.mc'..‘ :29). 1.. .1 .mfihanfl- "751- - . ...-.12 3110': L .113 .3 fl " ' _1! 7 ~ ‘ ‘ It‘llcnb'tua M 5 fire, 7 5m Mugv’,” ..av'. ‘. 0 How .‘1 ,-‘ ' 0‘ ' QGO§N99.:‘I Q‘Jdpm " \ ‘L .V :‘9: 1.1; r 1. . 773'.“ 1" J .vsw V ' 0.. .‘r . f‘ . -..£v’ 3.701891“ '1 ' .' -..-..-3 '7 (.1st (a .i rag); 5.73.»; me. > .._‘ \ovfimmfl) ‘ '- .812"; 3 10V .fl’CI . L . ‘J , . » t’j7 ." '1' 7417.357; M 1 9‘9”“ "T ' . 7 .1"?ka ‘ .. -‘. .‘. .7 T6 ”~53 one 5911.: '. M ((1, some: .s’arear m men J?‘ 719”: H ”mm. m g}; 5:51;? vsR T8113 )1 L .m ‘71. :38 (tn-3:10 mi- 1. .1. .M 75’“ "<-<.".~.'.7;.,.~a2.73_ L. -. 41.-V791 -- .7 ... 11:1} "axib-l'i'IN- 35 38571" them?) 2. F .rtrfirwfigk’ _ 1N Zarnefl.‘ (of {-8172 Y8r6~¢-"?lc 23* 3 .099? 562 men?) .1119. .1. w 1. 31:3): 1 ‘N nommnneq C) J .1003 :W .8 n1ofinoi31ATs‘ic‘V .3 f’ ams’oA (d1 Mia-SW! .0 R ,IetlsdausH (o) £01401 .‘S 9801.no1bet“1\oq.3.fl all 88b 32:3 #5. east 19713.83 m .M maul J. \ \‘1‘7'7 , ARE-"WTW'. L 11193377“ ‘ J -‘ ...7 y '. ,r- v ‘ 1 " 31334173 :1 _yg .1,- £35! . 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 312 Kittel, C. Introduction to Solid State Physics, 51h Ed.; John Wiley 87 Sons: New York, 1976. DD 413-416. Klepp, K. O. J. Less-Common Met. 1987, fig, 79-89. Zhang, X.; Kanatzidis, M. G. unpublished result. Temansky, R. J.; Balogh, D. w.; Paquette, L. A. J. Am. Chem. Soc. 1982, 1% 4503-4504. Eisenmann, B.; Katzer, H.; Schafer, H.; Weiss, A. Z. Naturforsoh. 1969, fig, 456-457. (a) Charnock, J. M.; Garner, C. D.; Pattrick, R. A. D.; Vaughan, D. J. J. Solid State Chem. 1989, 82, 279-289. (b) Wuensch, B. J. Z. Kristallogr. 1964, _1_1_9, 437-453. (c) Pauling, L.; Newmann, E. W. Z. Kristallogr. 1934, 8_8_, 54-62. (a) Gerl, H.; Eisenmann, B.; Roth, P.; Schafer, H. Z. Anorg. Allg. Chem. 1974, 407, 135-143. (D) Jumas, P. J. -C.; Ribes, M.; Maurin, M.; Philippot, E. Acta Crystallogr.1976, 32B, 444-448. Whangbo, M. -H.; Canadell, E. submitted for publication. (a) Kawai, S.; Ito, Y.; Kiriyama, R. Kobutsugaku Zasshi1972, 10(6), 487- 489. (b) Kiriyama, F1. Mem. Inst. Sci. Ind. Hes, Osaka Univ., 1977, E1. 23-38. (a) Tendeloo, G. V.; Amelinckx, 8. Acta Crystllogr. 1986, 4_2B, 121-130. (D) Hulliger, F. Structural Chemistry of Layer-Type Phases", Lévy, F., Ed.; D. Reidel Publishing Co: 1976; Vol 5, pp 228-229. (a) Pertlik, F. 2. Kristallogr. 1964, 169, 227-236. (D) Janner, A.; Dam, B. Acta Crystallogr.1989,45_A, 115-123. (c) Tunnel, 6.; Pauling, L. Acta Crystallogr. 1952, 5, 375-381. (d) Tendeloo, G. V.; Gregoriades, P.; Amelinckx, S. J. Solid State Chem. 1983, i0, 321-334. (e) Tendeloo, G. V.; Gregoriades, P.; Amelinckx, S. J. Solid State Chem. 1983, fig, 335- 361. (t) Schutte, W. J.; Boer, J. L. Acta Crystallogr. 1988, i4; 486-494. I." ‘At'.’ «CMIQ 4'". -U'l m, . seer 308 mm .mA .1. .A .1,th .6 ”:3 ~ I I'I7 "7‘73 - :77... 2277;..7 _3. A eairaw ; H Jemima ;.H 310901le is: ". ' ii " 4'11: f? 'J.1Btl‘760'.M .L 31 H M .«’.J"\ ’ "1:11.713 :8 .9“! M M '7 .- t'.v 2 _ , lips-’4 {13 869-189 .311, 9.3-«2 an») .. ‘-‘; l: 1W 7111... .3" '7r1£:~W-8{3,.H ,1‘9 '30": . J . Wind (3‘, c». C81 51%.. .7 3-... ‘,-.3. (4311371., 733.; 3‘- .low '1 L" .97 wafizzwnge 3' ‘7_ 29.17:... .H- V .0001.” SI 77'; .d-I‘J' ‘ I. ‘.. “7 7 ‘/ 071.8,18W‘x a) " 7 ".7! 17 2’1 -7 umzv‘lm (d) , ‘. . 7 7 1, 777.57 ; '7' .0 EO‘BDGOT (A) :7 . ...... «7. - 7 2y - v. '7; 9.274.111? 11 79131111114 (do) '3‘7' 3; 5‘ ‘ '_ ‘ .‘ ‘ "A fI!‘7' .14:qu WHO *- '14:;‘3". l 7 ‘7: 7-: . ' 7 ' .432" 7?“ .V. 4 ”1010919) 7: ; 67777.73: .9. 7 37: » new 73011312113” zyL-E-k y: :_ 7 {‘1 740:7? “2 .. 3391.1901va :3 00600:: 7' 103 1:77;; 5’87? 35‘3T7‘3f :77 7'73 . 8 mm 407*") 9181?. “L‘v‘ai': '1 E .7.A.,717:'.?7m~‘ 2‘ 881351109010; .V . .1 67227356 5393889: “'Ql.‘)‘719.‘-?.'f7.3£51.:- ..! 0.1-77:75:. w swung-2m .18 .3" mm 7308 men?) _mA LA .J,emups‘1:.w.0. C '27.: ..- p ‘ ‘_)*'\ , A“ lwifi'V‘y’ , 'L‘ '18.ng2 “H .19” ;. ‘ 7. 7' -"7 ' _"z J,1€‘7a'1“7.‘€' M ..L < ‘.T‘.J \ E; u a ”cit: emf M5” fl ‘ Y" I;‘ }‘7L':4 \3, Lzfi ‘8‘ guy we .3! a... L‘ '7 -("\I ‘p “ 4WD”; .3733 H "D ”F772 _ F'fi'ui 13') ('L: ”Q; .. J 777.7 .3. ..7‘7'w7’; .1qu 3.1.7071“ . 7 7n ~77 7 7- 5.77.77.77.- 77 mmm g 7;, . ;. _ 7' 27:7 .8 Jaws)! (n) 7‘ ‘7 ' 7779?! ”7 37:19.77”): ((3) ‘ ‘ I." r— ‘ fl 3 5" '17,, é) "'"i’DnOT “) - 7 7 7.1 .-: 7:247:11 WHUH (d) ‘ ,' ‘ ‘.‘.’ (‘E 1:177 9.1““- MM 3:.“ I" (d '3 )5“?qu (.) 7. ,. 3,343”? 1979731an 313A . .737... a 1597 momwfl ’7 ., 7 , 7 ,37‘ .745» .771; atrfii. 7377-73 ' 8 .xxaflim , “ ' ‘ 8171;777:7977: ’ asmsévcpmezv ‘7 < 7_. ;.' 7.3.7775: 19.3.- .,', 1,3905} .7 Vv' .SJIngba 0) JOE .3" .777 .II I "IWflififilflflmflifl‘!“fl‘fififlflfii‘ififlr HICHIGQN STATE UNIV. 18 L WWI III II II IWIII £31 2930104909 RRRIES WI 653