‘ V , 2.114161“... u‘\\\ L J 1‘ fiat-5” .‘ 'ovenou mes: ‘ 25¢ per day per ite- REIQRNLNG LIBRARY MTERIALS: Puco in book return to remove charge from circulation records STUDY OF MACROCYCLIC COMPLEXES IN NONAQUEOUS SOLVENTS BY Gholamhossein Rounaghi A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1980 ' kw / ._// ABSTRACT. STUDY OF MACROCYCLIC COMPLEXES IN NONAQUEOUS SOLVENTS BY Gholamhossien Rounaghi + .+ + , L1 , and Ag The complexation of the Na+, Cs+, Tl ion by the macrocyclic ligands 1,10-dithia-18-Crown-6 (DT18C6) and 1,4,7—trithia-12-Crown-4 (TT12C4) were studied by the multinuclear magnetic resonance technique in several nonaqueous solvents and their stabilities are compared with the complexes formed by the analogues polyethers 18-Crown- 6 and lZ-Crown-4. In all cases the substitution of the sulfur atoms for the oxygen atoms results in a substantial decrease in the stability of the complexes. In addition, the formation constants of the complexes varies significantly with the solvent. In general, the stabilities of the complexes vary inversely with the cationic solvating abilities of solvents as expressed by the Gutmann donicity scale. Comparison of the data obtained from 205T1, 23Na, 133Cs NMR studies in various solvents shows that the and stability of the appropriate complexes decreases in the order (mace-T1)+ > (DTlsce-Na)+ > (DT18C6-Cs)+ regardless Gholamhossien Rounaghi of the solvent in which the complexation takes place. The formation constants for the complexation of the Cs+ ion with large macrocyclic crown ethers, dibenzo-27- Crown-9 (DBZ7C9), and dibenzo-24-Crown-8 (DBZ4C8) were studied at different temperatures in various organic sol- vents using the Cesium-133 NMR technique. Enthalpies (AH:) and entropies (Asg) of the complex formation were calculated from the Van't Hoff relationship, and it was found that both quantities are quite sensitive to the sol- vent. In all cases the complexes were enthalpy stabilized but entropy destabilized (As: < O). The origins of the enthalpies and entropies of complexation are discussed in terms of structural features (flexibility) of the ligands and of the solvation effect, and both parameters show inter- esting and consistent changes as a function of the ligand cavity size and number of binding sites in the polyether rings. The thermodynamics of the complexation of the Cs+ ion by the macrocyclic polyethers, DBZ7C9, and DBZ4C8 was studied in dimethylformamide-acetonitrile binary systems. Experimental values of AGE, AHE, and A53 again indicate that these quantities are quite sensitive to the solvent composition. In addition, in all cases the complexes are enthalpy stabilized but entropy destabilized. Complex formation by the cesium ion with cryptand- 222 (C222) was investigated in acetone-dimenthylsulfoxide, Gholamhossien Rounaghi acetonitrile-dimethylsulfoxide, propylene carbonate- dimethylsulfoxide, and propylene carbonate-dimethyl- formamide binary mixtures. The formation constant, Kf, of the (C222-Cs)+ complex in these systems increases with in— creasing amounts of acetone, acetonitrile, and propylene carbonate in the respective binary mixtures, indicating a poorer solvating ability of these solvents for the Cs+ ion and the ligand. In all cases the limiting chemical shifts of the cesium-cryptate are dependent on the solvent com- positions, indicating that the Cs+ ion is only partially insulated (by the ligand) from the solvent. Cesium-133 NMR measurements were performed on competitive solvation of the Cs+ ion in dimethylsulfoxide— water, acetonitrile-water, and methanol-water binary sol- vents. The geometric equilibrium constant, Kl/n, and the free energy of preferential solvation were obtained using the Convington equation. The order of solvating ability of these solvants was found to be dimethylsulfoxide > water ~ methanol > acetonitrile. The complexing ability of the macrocyclic ligand, cyclo-(tetraethylene-glycol—Z,6-pyridine dicarboxate) with the Na+ and Cs+ ion was studied in various nonaqueous sol- vents. This ligand forms a very stable complex with the Na+ ion in acetone, acetonitrile, and propylene carbonate solutions. In the case of the Cs+ ion the data clearly show the formation of both 1:1 (metal ion:ligand) and 1:2 Gholamhossien Rounaghi (sandwich) complexes in propylene carbonate and dimethyl- formamide solutions. The results obtained from the chemical shift measurement of the 205T1 resonance on complexing of Tl+ ion by benzo-lS-Crown-S in nonaqueous solvents demonstrate the weaker complexing ability of this ligand than that of lB-Crown-G for thallium ion. The complexation reaction between the Tl+ ion and C222 was studied in acetone, acetonitrile, and dimethyl- sulfoxide solutions using 205 T1 NMR technique. Cryptand- 222 forms a very stable complex with Tl+ ion in these sol- vents. The rate of exchange between the free Tl+ ion and the complexed-Tl+ ion is fast on the NMR time scale; there- fore, two 2OSTl signals are observed as long as the mole ratio of the C222/Tl+ is between zero and unity. ACKNOWLEDGMENTS The author wishes to thank Professor Alexander I. Popov for his guidance, counseling and encouragement throughout this study. He also wishes to thank Professor Michael J. Weaver for his helpful suggestions as second reader. The financial aids of the peOple of Iran, as administered by the Ministry of Science and Higher Education is acknowledged. Gratitude is also extended to the Department of Chemistry, Michigan State University and the National Science Foundation for financial aid. I would like to thank all the members of the laboratory of Dr. A.I. Popov for their friendship and discussion, and Mr. Wayne Burkhardt for his effort in keeping the NMR spectrometer in operating condition. Finally, deep appreciation is extended to my mother, my wife, and my sisters for their patience, consideration, and encouragement throughout this study. To them and to our son, Ali, I dedicate this thesis. ii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O O O O O O 0 LIST OF FIGURES . O O O C O O O O I O O O O O 0 Chapter I II HISTORICAL REVIEW . . . . . . . . . . . A. B. I?! U 0 '71 Introduction . . . . . . . . . . . . Macrocyclic Crown Ethers and Their PrOperties . . . . . . . . . . . . a. Cation Selectivity and Complex Stability . . . . . . . . . b. Structure and Stiochiometry of Alkali Cation Crown Ether Complexes . . . . . . . . . . . Sulfur-Oxygen Mixed Donor Macrocylic Ligands . . . . . . . . . . Complexing of Metal Cations by Macro- -polycyclic Ligands- Cryptands . Thermodynamics of Cation—Macrocyclic Interaction . . . . . . . . . . . . Nuclear Magnetic Resonance . . . . . a. Introduction . . . . . . . . . .. b. Chemical Shift Measurements of Alkali Metal Nuclei . . . . . . c. Thallium-ZOS NMR Measurement . . EXPERIMENTAL PART . . . . . . . . . . . WINDOW? Synthesis and Purification of Ligands Salts . . . . . . . . . . . . . . . Solvents . . . . . . . . . . . . . . Instrumental Measurements . . . . . Sample Preparation . . . . . . . . . Data Handling . . . . . . . . . . . iii Page vi xii 13 17 20 26 31 31 32 36 4O 41 43 44 45 49 50 Chapter III IV VI MULTINULCEAR MAGNETIC RESONANCE STUDY OF + . COMPLEXATION OF Na+, Cs+, Tl , L1+, AND Ag+ IONS BY THIA-CROWN AND CROWN ETHERS IN NONAQUEOUS SOLVENTS . . . . . . . . . . A. Introduction . . . . . . . . . . . . . B. Results and Discussion . . . . . . . C. Comparison of the Results . . . . . THERMODYNAMIC STUDIES OF THE COMPLEXATION OF DIBENZO-27-CROWN-9 AND DIBENZO-24-CROWN- 8 WITH CESIUM ION IN VARIOUS NONAQUEOUS SOLVENTS . . . . . . . . . . . . . . . . . A. Introduction . . .+. . . . . . . Complexation of Cs IOn by+ DB27C9 . . DB24C8 Complex with the Cs+ Ion . . . Comparison of the Thermodynamic Parameters . . . . . . . . . . . . . . DOB? CESIUM-133 NMR STUDY OF THE CESIUM ION COMPLEXATION BY CROWN ETHERS AND CRYPTAND- 222 IN MIXED NONAQUEOUS SOLVENTS . . . . . A. Introduction . . . B. Complexation of the .Cs+; Ion by C222. . C. Complex Formation by Cs+ Ion with DB24C8 in DMSO/AC and PY/MeOH Binary Systems. . . . . . . . . . . . . . . . D. Thermodynamic Parameters for Com- plexing of Cs+ Ion by DBZ7C9 and DBZ4C8 in DMF/AN Binary Solutions. . . MISCELLANEOUS . . . . . . . . . . . . . . A. Ionic Solvation of Cs+ Ion in Mixed Solvents . . . . . . . . . . . . . B. Complexation of the T1+ Ion by C222 . C. Complexation Study of the Na+ and Cs+ Ions with Cyclo-(tetraethylene- glycol- 2, 6-pyridine dicarboxylate) in Various Solvents . . . . . .+. . . . . . . . D. Complexation of the T1 Ion by BlSCS . E. Thermodynamic Study of Complexation of DB21C7 with the Cs+ Ion in DMF and PC . . . . . . . . . . . . . . . . iv Page 51 52 53 79 86 87 88 117 125 129 130 130 146 166 185 186 194 197 204 208 Page APPENDIX 0 0 O O O O O o O 0 I O O 0 O O 0 O O O 0 217 REFERENCES 0 o o o o o o o o o o o 0 .V o o o o o o 222 Table 10 11 12 LIST OF TABLES Page Diameter in A° of macrocyclic crown ethers and some univalent cations. . . . . 6 Log Kf values for reaction of lSCS and thia-lSCS with some univalent cations in water 0 o I o o o o a o o o o o O O 0 0 18 Comparison of stability constants of metal cation complexes of macrocyclic ligands with their open chain analogs . . . . . . 29 Nuclear properties of alkali and thallium nuclei . . . . . . . . . . . . . . . . . . 33 Diamagnetic susceptibility correction on DA-60 for various solvents . . . . . . . . 47 Mole ratio-chemical shift data for sodium tetraphenylborate in the presence of DT18C6 in various solvents . . . . . . . . 54 Structure, Gutmann donor number, and dielectric constant of certain solvents . 61 Mole ratio-chemical shift data for CsSCN in the presence of DT18C6 in various nonaqueous solvents . . . . . . . . . . . 63 Log of formation constants and the limit- ing chemical shifts of (DTl8C6-Na)+ complex in various solvents at 32 i 1°C . . . . . 66 Log of stability constants and the+limit- ing chemical shifts of (DT18C6-Cs) complex in various solvents at 32 : 1°C . . . . . 66 Mole ratio-chemical shift data for TlClO4 in the presence of DT18C6 in various nonaqueous solvents . . . . . . . . . . . 67 Mole ratio-chemical shift data for 18C6 complex with Tl+ ion in various solvents . 71 vi Table 13 14 15 16 17 18 19 20 21 22 23 24 25 Stability constants of thallium (1) complexes with DT18C6 and 18C6 in various nonaqueous solvents . . . . . . . . . . . . Mole ratio-chemical shift data for LiClO4 in the presence of TT12C4 in various solvents at 30°C . . . . . . . . . . . . Hydrogen chemical shift data for TT12C4 in the presence of Ag+ ion in various solvents Stability constants of the sodium, cesium, and thallium (I) complexes with DT18C6 and 18C6 crown ethers in various solvents . Stability constants of (Lio12C4)+ complex in different solvents . . . . . . . . . . Mole ratio-chemical shift data for CsSCN (0.005M) in the presence of DBZ7C9 in NM at various temperatures . . . . . . . . . . Mole ratio-chemical shift data for CsSCN (0.005M) in the presence of DB27C9 in DMF at different temperatures . . . . . . . . . Mole ratio-chemical shift data for CsSCN (0.005M) in therpresence of DBZ7C9 in PC at various temperatures . . . . . . . . . . Mole ratio-chemical shift data for CsSCN (0.005M) in the presence of DB27C9 in AN at various temperatures . . . . . . . . . Mole ratio-chemical shift data for CsSCN (0.005M) in the presence of DBZ7C9 in AC at different temperatures . . . . . . . . . Mole ratio-chemical shift data for CsSCN (0.05M) in the presence of DBZ7C9 in FY at various temperatures . . . . . . . . . . Mole ratio-chemical shift data for complexation of Cs+ by DB27C9 in methanol at various temperatures . . . . . . . . . . Cesium-133 data for complexation of Cs+ ion byDBZ7C9 in methanol at various tem- peratures . . . . . . . . . . . . . . . . . vii Page 73 74 78 81 85 91 92 93 94 95 96 97 98 Table 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Mole ratio study of DB27C9 with CsSCN (0.005M) in DMSO at 30°C 0 o o o o o o 0 Formation constant of the (DB27C9-Cs)+ complex in various solvents at 30°C . . . Formation constants of (DBZ7C9-Cs)+ complex in various solvents at different temperatures . . . . . . . . . . . . . . . Thermodynamic parameters for (DB27C9-Cs)+ complex in various solvents . . . . . . . Cesium-133 Chemical shift data for com- plexation of Cs+ ion by DB24C8 in DMF at various temperatures . . . . . . . . . . . Cesium-133 chemical shift data for com- plexing of Cs+ ion by DB24C8 in PC at various temperatures . . . . . . . . . . . Stability constants of (DBZ4C8-Cs)+ complex in DMF and PC at various tem- peratures I O O O O O O O O 0 O O O O 0 Thermodynamic quantities for (DB24C8 Cs)+ complex in DMF and PC . . . . Comparison of thermodynamic quantities for complexation of Cs+ ion by DB27C9 and DBZ4C8 in various nonaqueous solvents Mole ratio-chemical shift data for CsSCN (0.005M) in the presence of C222 in DMSO/AC systems at 32°C . . . . . . . . Mole ratio study of C222 complex with 0.005M CsSCN in AN/DMSO systems . . . . . Mole ratio study of C222 complex with 0.005M CsSCN in PC/DMSO binary mixtures at 32°C 0 O O O O O O O O O O O O O O O O Mole ratio-chemical shift data for CsSCN (0.005M)131the presence of C222 in DMF/PC systems at 32°C . . . . . . . . . . Log of formation constants and limiting chemical shifts for (C222 Cs)+ complex in AC and DMSO and their mixtures at 32 i 1°C viii Page 99 100 112 116 118 119 122 124 126 131 133 134 135 143 Table Page 40 Log of stability constants and limiting chemical shifts for (C222-Cs)+ complex in AN/DMSO and PC/DMSO systems at 32°C . 144 41 Log of stability constants and limiting chemical shifts for (C222-Cs)+ complex in PC and DMF and their mixtures . . . . 145 42 Mole ratio-chemical shift data for CsSCN (0.005M) in the presence of DB24C8 in AC and DMSO and their mixtures at 32°C I O O O I O O I O O O O O O O O O O 147 43 Mole ratio-chemical shift data for CsSCN (0.005M) in the presence of DBZ4C8 in MeOH/PY systems at 32°C . . . . . . . . 149 44 Formation constants of (DBZ4C8-Cs)+ complex in DMSO-AC mixed systems at 32 : 1°C 0 O O O O O O O O O I O O O O O 156 45 Formation constants of (DBZ4C8~Cs)+ complex in MeOH/PY binary systems (at 32 : 1°C) 0 O O O O O O O O O O O O O O 157 46 Stimated formation constants of (C222-Cs)+ and (DB24C8.Cs)+ complexes in various compositions of AC/DMSO binary solutions . . . . . . . . . . . . . . . 164 47 Mole ratio-chemical shift data for com- plexation of DBZ7C9 with CsSCN (0.005M) in 10% AN + 90% DMF at various tem- peratures . . . . . . . . . . . . . . . 168 48 Mole ratio-chemical shift data for com- plexation of DB27C9 with CsSCN (0.005M) in 30% AN + 70% DMF at different temperatures 169 49 Mole ratio-chemical shift data for DBZ7C9 complex with 0.005M CsSCN in 70% AN + 30% DMF at various temperatures . . 170 50 Mole ratio-chemical shift data for com- plexation of DB24C8 with CsSCN (0.005M) in 30% AN + 70% DMF at various tempera- tures O O O O O O O O O O O O I O O O O 171 ix Table Page 51 Mole ratio-chemical shift data of DBZ4C8 complex with CsSCN (0.005M) in 70% AN + 30% DMF at different temperatures 172 . + 52 Formation constants of (DBZ7C9-Cs) com- plex in AN/DMF binary solvents at various temperatures . . . . . . . . . . . . . . . 178 53 Formation constants of (DBZ4C8-Cs)+ com- plex in AN/DMF binary solvents at different temperatures . . . . . . . . . . . . . . . 179 54 Thermodynamic parameters for (DB27C9-Cs)+ and (DB24C8-Cs)+ complexes in DMF and AN and their binary mixtures . . . . . . . . 181 55 Cesium-133 chemical shifts of cesium salt solutions in mixed solvents . . . . . . . 187 56 I.S.P. data, Kl/n, and AG°/n for Cs+ ion in the binary mixed solvents . . . . . . . 192 57 Thallium NMR study of TlClO4 with C222 in different solvents . . . . . . . . . . 195 58 Mole ratio study of sodium tetra- phenylborate (0.02M) in the presence of cyclo (tetraethylene-glycol-2,6-pyridine dicarboxylate) in various solvents at 32 : 1°C . . . . . . . . . . . . . . . . . 199 59 Mole ratio study of cyclo (tetraethylene- gycol-2,6-pyridine dicarboxylate) complex with Cs+ ion in various solvents . . . . . 202 60 Chemical shift-mole ratio data for com- plexation of T1+ ion by B15C5 in various solvents . . . . . . . . . . . . . . . . . 205 61 Mole ratio study of DBZlC7 complex with 0.005M CsSCN in DMF at different tem- peratures . . . . . . . . . . . . . . . . 209 62 Mole ratio study of DBZlC7 complex with 0.005M CsSCN in PC at different temper- atures O O O O O O C O O O O O O O O O O O 210 63 Stability constants of (DBZlC7-Cs)+ com- plex in DMF and PC at various tempera- tures O O O O O O O O O O O O O O I O O O 213 X Table Page 64 Thermodynamic parameters for (D321c7-Cs)+ complex in DMF and PC . . . . 215 xi LIST OF FIGURES Figure Page 1 Structure of some Synthetic Macrocyclic Ligands . . . . . . . . . . . . . . . . . 5 2 Variation of Log K vs. Cation Radius for Interaction of 15C5, 18C6 and 21C7 with Univalent Cation in Methanol at 25°C . . 7 3 Structure of Cryptand-222 and Cryptand- 22C8 O O O O O O O O O O O O O O O O O O 12 4 Crystalline Structure of (KI-DB30C10) . . 15 5 Structure of DBZ4C8-(K+)2 . . . . . . . . 15 6 Sandwich Structure of (B15C5)-K+ Complex 16 7 Crystalline Structure of 1,4,7-trithia-12- crown-4 o o o o o o o o o o o o o o o o o 19 8 Structure of Cryptands (with internal diameter) . . . . . . . . . . . . . . . . 21 9 ‘ Conformations of Cryptand-222 . . . . . . 23 10 The Structures of (a) Cylindrical, and (b) Spherical Macrotricyclic Cryptands . 24 11 Successive Formation of Mononuclear and Binuclear Cryptate of Cylindrical Macro- cyclic Cryptand . . . . . . . . . . . . . 25 12 Intramolecular Cation Exchange Process in Mononuclear Complexes of Cylindrical Cryptand . . . . . . . . . . . . . . . . 25 13 Chemical Shifts of 23Na as a Function of DT18C6/Na+ Mole Ratio in Various Solvents 55 14 Sodium-23 Chemical Shifts vs. DTlace/Na+ Mole Ratio in AC and DMF . . . . . . . . 56 15 Cesium-133 Chemical Shifts vs. DTlSCS/Cs+ Mole Ratio in Various Solvents . . . . . 64 xii Figure Page 16 Thallium-ZOS Chemical Shifts vs. D18C6/T1+ Mole Ratio in Various Solvents 68 17 Thallium-205 Chemical Shifts vs. 18C6/T1+ Mole Ratio in Various Solvents 72 18 Lithium-7 Chemical Shifts vs. TT12C4/Li+ Mole Ratio in Nonaqueous Solvents . . . . 77 19 Structural Formula of 1,4,7-trithia—12— CI'OWD’4 o o o o o o o o o o o o o o o o o 79 20 The Variation of the Proton Chemical Shift of the Three Different Protons of TT12C4 with the Ag+/TT12C4 Mole Ratio in DMSO-d6 80 21 Cesium-133 Chemical Shifts vs. DB27C9/Cs+ Mole Ratio in Various Solvents . . . . . 101 22 Cesium-133 Chemical Shifts vs. DB27C9/Cs+ Mole Ratio in Various Solvents . . . . . 102 23 Cesium-133 Chemical Shifts as a Function of DB27C9/Cs+ Mole Ratio in DMF at Different Temperatures . . . . . . . . . 103 24 Cesium-133 Chemical Shifts vs. D1327C9/Cs+ Mole Ratio in PC at Different Temperatures 104 25 Cesium-133 Chemical Shifts vs. DB27C9/Cs+ Mole Ratio in NM at Various Temperatures 105 26 Cesium-133 Chemical Shifts vs. DBZ7C9/Cs+ Mole Ratio in AN at Various Temperatures 106 27 Cesium-133 Chemical Shifts vs. DB27C9/Cs+ Mole Ratio in AC at Various Temperatures 107 28 Cesium-133 Chemical Shifts vs. D327C9/Cs+ Mole Ratio in PY at Various Temperatures 108 29 Cesium-133 Chemical Shift vs. DBZ7C9/Cs+ Mole Ratio in MeOH at Various Temperatures 109 30 Cesium-133 Chemical Shift vs. DB27C9/Cs+ Mole Ratio in MeOH at Various Temperatures 110 31 A Plot of 1n K vs. % for the Complexation of Cs+ Ion with DB27C9 in Various Solvents 114 xiii Figure Page 32 Cesium-133 Chemical Shifts vs. DBZ4C8/Cs+ Mole Ratio in DMF at Various Tempera- tures O O 0 O O O O O O O O O I O O O I O 120 33 Cesium-133 Chemical Shifts vs. D1324C8/Cs+ in PC at Various Temperatures . . . . . . 121 34 A Plot of In K vs. % for the Complexation of Cs+ Ion with DB24C8 and DB27C9 in PC and DMF o o o o o o o o o o o o o o I Q o 123 35 Cesium-133 Chemical Shifts vs. C222/Cs+ Mole Ratio in DMSO/AC Binary Mixtures . . 138 36 Cesium-133 Chemical Shifts vs. C222/Cs+ Mole Ratio in DMSO/AC Binary System . . . 139 37 Cesium-133 Chemical Shifts vs. C222/Cs+ Mole Ratio in AN/DMSO Binary Solutions . 140 38 Cesium-133 Chemical Shifts as a Function of C222/Cs+ Mole Ratio in PC/DMSO Binary Mixtures I O O O O O O O O O O O O O O 0 141 39 Cesium-133 Chemical Shifts vs. C222/Cs+ Mole Ratio in DMF/PC Binary Solutions . . 142 40 Cesium-133 Chemical Shifts vs. D1324C8/Cs+ Mole Ratio in AC/DMSO Systems . . . . . . 152 41 Cesium-133 Chemical Shifts vs. DB24C8/Cs+ in AC/DMSO Systems . . . . . . . . . . . 153 42 Cesium-133 Chemical Shifts as a Function of DB24C8/Cs+ Mole Ratio in Methanol- Pyridine Binary Solutions . . . . . . . . 155 43 Variation of the Stability Constant of (C2220Cs)+ Complex with Solvent Composi- tion in Different Binary Solutions . . . 159 44 Variation of the Formation Constant of (C222-Cs)+ Complex with PC Content in PC/DMF Binary Mixtures . . . . . . . . . 160 45 Variation of the Stability Constant of (DE24C8-Cs)+ Complex with the Solvent Composition in MeOH/PY and DMSO/AC Binary Solutions . . . . . . . . . . . . . . . . 162 xfil Figure Page 46 Variation of the Stability Constants of (a) (DB24C8-Cs)+ and (b) (C222.CS)+ Complexes with the Reciprocal of the Dielectric Constant of the Medium in AC/DMSO Binary Solutions . . . . . . . . . 165 47 Cesium-133 Chemical Shifts vs. D1327C9/Cs+ Mole Ratio in 10% AN + 90% DMF at Various Temperatures . . . . . . . . . . . . . . . 173 48 Cesium-133 Chemical Shifts as a Function of DEZ7C9/Cs+ Mole Ratio in 30% AN/70% DMF at Various Temperatures . . . . . . . . . 174 49 Cesium-133 Chemical Shifts vs. DB27C9/Cs+ Mole Ratio in 70% AN/30% DMF at Various Temperatures . . . . . . . . . . . . . . . 175 50 Cesium-133 Chemical Shifts vs. DB24C8/Cs+ Mole Ratio in 30% AN + 70% DMF at Various Temperatures . . . . . . . . . . . . . . . 176 51 Cesium-133 Chemical Shifts vs. DB24C8/Cs+ Mole Ratio in 70% AN/30% DMF at Various Temperatures . . . . . . . . . . . . . . . 177 52 Van‘t Hoff Plots for Complexation of the Cs+ Ion by DB27C9 and DB24C8 in AN/DMF Binary Solutions . . . . . . . . . . . . . 180 53 Cesium-133 Chemical Shifts vs. Composition of AN in HZO-AN Binary Mixtures . . . . . 188 54 Cesium—133 Chemical Shifts vs. Composition of MeOH in HZO-MeOH Binary Solutions . . . 189 55 Cesium-133 Chemical Shift as a Function of DMSO Content in HZO-DMSO Binary Solutions 190 56 Thallium-205 NMR Spectra of (C222-T1)+ Complex at Various C222/T1+ Mole Ratios in Dimethylsulfoxide Solutions . . . . . . 196 57 Thallium-ZOS Chemical Shifts of Both, Free and Complexed T1+ Ion as a Function of C222/T1+ Mole Ratio in Various Solvents 198 58 Sodium-23 Chemical Shifts vs. Cyclo- (tetraethylene-glycol-Z,6-pyridine dicarboxylate)/Na Mole Ratio in Various Solvents . . . . . . . . . . . . . . . . . 200 XV Figure Page 59 Cesium-133 Chemical Shifts vs. Cyclo— (tetraethylene-glycol-Z,6 pyridine dicarboxylate)/Cs+ Mole Ratio in Various Solvents . . . . . . . . . . . . . . . . . 203 60 Thallium-ZOS Chemical Shifts vs. 15C5/T1+ Mole Ratio in Various Solvents . . . . . . 206 61 Cesium-133 Chemical Shifts vs. DB21C7/Cs+ Mole Ratio in DMF at Various Temperatures 211 62 Cesium-133 Chemical Shifts vs. D321C7/Cs+ Mole Ratio in PC at Various Temperatures 212 63 Van't Hoff Plots of 1n Kf vs. % for (DB21C7-Cs)+ System in PC and DMF . . . . 214 xvi LIST OF ABBREVIATIONS I. LIGANDS 18C6 18-Crown-6 DT18C6 1,10-dithia-18-Crown-6 TT12C4 1,4,7-trithia-12-Crown—4 DB21C7 Dibenzo-Zl-Crown-7 DB24C8 Dibenzo-24—Crown-8 DBZ7C9 Dibenzo-27-Crown-8 BlSCS Monobenzo-lS-Crown-S C222 Cryptand-222 II. SOLVENTS NM Nitromethane AN Acetonitrile PC Propylene carbonate AC Acetone MeOH Methanol DMF Dimethylformamide DMSO Dimethylsulfoxide PY Pyridine TMG Tetramethylguanidine THF Tetrahydrofuran xvii CHAPTER I HISTORICAL REVIEW A. Introduction Although alkali metal cations play important role both in chemistry and in biology, the complexes of alkali metal ions were not considered by chemists for many years, and most chemists assumed that such complexes are not stable. In recent years, however, the coordination chemistry of alkali metal ions has mainly developed by the synthesis of crowns by Pedersen (1) and cryptands by Lehn and co-workers (2). Macrocyclic polyethers (crowns) and macrobicyclic polyethers (cryptands) have been shown to form very stable complexes with the alkali and alkaline earth metal cations. These complexes can be used as models for investigation of ion transport through membranes in biological systems (3). Under ordinary conditions, the movement of electrolytes across the membranes would be eXtremely small because of low solubility of electrolytes in the organic layer. How- ever, by proper choice of cyclic polyether complexing agent one may be able to achieve the selective cation complexa- tion, and rapid and reversible reaction of the cation with the cyclic polyether followed by diffusion of the soluble cation-cyclic polyether through the membrane. A large number of physicochemical techniques have been used to study the thermodynamic and kinetic aspects 2 of macrocyclic complexation (4). Of the various methods used for such studies,a1kali metal NMR has been shown to be a very sensitive and powerful technique, since the chemical shifts of alkali nuclei are very sensitive to the immediate environment of an alkali ion in solution. B. Macrocyclic Crown Ethers and Their Properties a. Cation Selectivity and Complex Stability Naturally occurring macrocycles were shown to be capable of actively transporting metal ions across mem- branes, beginning with valinomycin in 1964 (5). Macro- cyclic polyethers, which are similar to the antibiotic ligands, both in structure and in their ability to form stable complexes with alkali cations were first reported in 1967 by Pedersen (1). After the first publication by Pedersen, the chemistry of crown ethers has developed rapidly, and in many areas of chemistry and biology these compounds have started to play important roles. They have been successfully used for complexation, where they show a high degree of cation selectivity (6). The crown ethers are uncharged molecules in which donor oxygens are arranged in a ring, usually each separated -CH -. The 2 2 rings are of different sizes, containing from 4 to 20 from the other by two methylen groups, -CH oxygen atoms. Many of those (macrocyclic polyethers) con- taining five to ten oxygen atoms form stable complexes with alkali and alkaline earth cations as well as some of the transition metal ions. Pedersen (7) found that the most favorable linkage for complexation process is -0-CH2-CH -0- followed by -0-(CH2) -0- and becomes in- 2 3 effective beyond -0-(CH -0-. 2)4 The structural formula of several synthetic crown ethers are shown in Figure 1. Since the names of these compounds are too cumbersome for use, the abbreviated names suggested by Pedersen are usually used for their identification. For example the name of compound (I) is 18-Crown-6 (or 18C6), in which the first number refers to the total number of atoms in the polyether ring, while the second number stands for the number of oxygen or other donor atoms. When sulfur atoms are substituted for the oxygen atoms a ”thia" prefix is used. Therefore compound (II) is called 1,10—Dithia-18-Crown-6 (or DT18C6). The most important property of the macrocyclic polyethers is their tendency to form very stable complexes with alkali ions with the stabilities that are comparable to those of transitibn metal complexes. The metal ion- crown ether complexes are normally considered to be formed by an electrostatic attraction between the cation and the dipolar end of the -C-0 dipoles. It has been shown that the cation selectivity and complex stability of macrocyclic ligands depend on several important parameters. The more obvious ones being: lS-CROHN-B (18C6) (1) 6°C” 0 0 tom.) DIBENZO-Zl-CROHN-7 (D321C7) (III) (\Oflo’} @100 .8 Lo 0—) K/OxJ DIBENZO-27-CROHN-9 (DB27C9) (V) 1,10-DITHIA-18-CROWH-6 (DT18C6) (II) DIBENZO-ZQ-CROWN-S (DB24C8) (IV) b/"\ C s o c a C! 1.4.7-TRITHIA-12-CROHN-4 (TT12C4) (VI) Figure 1. Structure of Some Synthetic Macrocyclic Ligands. 1. The relative sizes of cation and the macrocyclic cavity: The stability of the polyether complexes depends primarily upon how well the cation fits into the polyether ring. The prefered cation being that which has the best fit into the cavity. The size selectivity is determined by several types of primary effects such as: electrostatic and van der Waals attraction and repulsion between metal and ligand, ligand conformational changes, desolvation of the metal ion and the ligand, solvation of the metal ion- 1igand complex, and the nature and number of binding sites. The ionic diameters and the sizes of cavities of some of the crown ethers are listed in Table 1, and the relationship between the log of stability constant and Table 1. Diameter inZV’of macrocyclic crown ethers and some univalent cation (in crystals) Cation Cation (a) Crown Crown ether diameter ethers diameter Li+ 1.20 All lS-Crown-5(b) 1.7 - 2.2 Na+ 1.90 All 18-Crown-6(b) 2.6 - 3.2 K+ 2.66 A11 21-Crown-7(b) 3.4 - 4.3 Rb+ 2.96 Larger than 21- Crown-7Ia) over 4 Cs+ 3.34 Ag+ 2.52 Tl+ 2.80 (a) (b) Reference 1 Reference 8 cation radius for interaction of macrocyclic crown ethers, 15C4, 18C6, and 21C7 with some of the univalent cations in methanol is shown in Figure 2. j I 1 1 I 6.0‘ 2107 5.0'4 18C6 4.0- a: 1: )fi ' 8’ 304 .I 2.0~ ISCS 1.0-i I I I I I Na" Ag‘ K+ Rb’ Cs’ I' 1 T I 1 1.00 1.20 1.40 1.60 1.80 CaIion Radius (A) Figure 2. Variation of Log K vs. Cation Radius for Interaction of 15C5, 18C6, and 21C7 with Univalent Cation in Methanol at 25°C (9). Although the fitness of the ligand ring cavity size and the cation size is an important factor in determining the stability of macrocyclic complexes, however, in some cases the size correspondence is not the predominant factor in determining selectivity,and other factors such as cation solvation energies, the solvation energy of the ligand and of the complex, and the formation of complexes of stiochiometry other than 1:1 may govern the selectivity of macrocyclic complexes. For example it has been shown (9) that in methanolic solutions K+ is bound more strongly by lS-Crown-S than Na+ even though Na+ ion has a better fit for the ligand cavity than K+ ion. It should be noted that the fit of the alkali cations in the ligand cavity can not explain the selec- tivities of large crown ethers (24-Crown-8 and larger) because their cavities are all larger than the largest alkali cation, Cs+ (see Table 1). However, large stability constants still can be found, even when the cavity is much bigger than the metal cation. For example, the log of stability constants for complexes of K+ ion with 18-Crown- 6, 21-Crown-7, 24-Crown-8, 30-Crown-10, and 60-Crown-20 in methanol have been reported by Frensdorff (10) to be 5.0, 4.3, 3.5, 4.6, and 3.9 respectively. Apparently, the relatively large number of oxygen ethers and probably the flexibility of the ligand compensates for a decrease in stability when the size of the cavity is much larger than that of the cation. 2. Solvation power of the medium The complexation process between the ligand and the metal cation is represented by the following general equation: (metaln+)sx + (polyether)Syt=?(polyether - meta1)n+sz + (x + y - z)S in which S istflmasolvent molecule, x, y, and z are the solvation number of metal cation, ligand, and complex respectively. To what extent the complexation takes place obviously depends greatly upon the strength of the solvation of all species involved. In complexed species, the macrocyclic polyether replaces solvent molecules in the solvation sphere of the cation, therefore, the cation must lose the solvent molecules from its coordination sphere before it can accept a ligand. It should be stressed, however, that not only the metal ion, but also the ligand is solvated. The formation of a complex in solution is therefore, not only a competition for the metal ion between the ligand and the solvent molecules, but also a competition for the ligand between metal ion and the solvent molecules. Hence, the formation of the complex for a particular ion may be minimized, or even prevented, if the metal cation and the ligand are strongly solvated by the solvent. In addition, selectivity for certain cations over others may be altered according to the nature of the solvent. 10 The solvation of the ligand and metal ion can be influenced by the donor ability and dielectric constant of the solvents as well as the stereochemistry and size of the solvent molecules. Popov and co-workers have studied extensively the complexation of alkali cations by macrocyclic ligands in a wide variety of nonaqueous solvents using alkali metal NMR technique (ll-l7). Their studies revealed that the stability of cation-macrocyclic complexes depends not only on the relative size of the ion and macrocyclic cavity but to a large extent on the nature of the solvent in which complexation takes place, i.e., on the solvent- ion, solvent-ligand, and ion-ion interactions. Arnett and Moriarity (18) who studied complexation of alkali cations with dicyclohexo-lB-Crown-6 in various solvents reported that the stabilities of the complexes of large cations are less affected by solvent than those of smaller ones. 3. Type(s) of binding sites in the ring Various modifications have been made to the basic crown ether structure to enhance the selectivity of these ligands and the stability of complexes formed. Among these modifications are the substitution of ligand donor atoms such as sulfur and nitrogen for one or several of the polyether oxygens in the polyether ring. 11 Substitution of nitrogen or sulfur for oxygen re- duces affinity for alkali ions but greatly increases the complexing power of these ligands for silver ion, because the covalent bonding plays a role in the silver ion com- plexes of polyethers containing nitrogen or sulfur (10). In general, for a ligand with a donor atom such as 0, N, or S, the stability constant for alkali metal cations follows the trend, 0 > NR > NH > S. An opposite stability trend may be expected when the cation-ligand bond has appreciable covalent Character as in the case of Ag+...N or Ag+...S interactions. Replacement of all of the oxygen donor sites in a macrocyclic ligand with sulfur causes the ligand completely loses its complexing ability for the alkali and alkaline earth cations (l9). 4. Number of donor atoms The number of binding sites certainly play a role in the complexing properties of macrocyclic polyethers. The polyether must have the sufficient number of the donor atoms in order to get the large complexation energy which is required for the desolvation step of the cation and ligand. Although little quantitative work has been done to investigate the effect of varying the number of donor atoms in the polyether ring, however, the relationship between the number of oxygen atoms in the crown ligand and the complex stability can be pointed out. 12 Cram et a1. (20) reported that 18-Crown-5 is a much poorer complexing agent for t-butly amonium ion than 18- Crown-6. Studies of complexation reactions of C222 and C22C8 (Figure 3) with Na+ and K+ in methanolic solutions by Lehn and co-workers showed that the C222 complexes of Na+ and K+ ions are much more stable than the complexes formed between C22C8 with these ions (21). Both ligands have about the same cavity size but C22C8 has two fewer oxygen sites. A corresponding decrease in the stability by a factor of 104-105 is observed. 0 o N/\x/\/x\n 0\ /cr\\~/’J X = 0 C222 X = CH2 C22C8 Figure 3. Structure of Cryptand-222 and Cryptand~22C8. 5. The electric charge of the cation In general, with a similar cation radius, a bivalent ion has a higher stability constant than a Imonovalent ion. For example, it has been shown (6) that complexes of bivalent cations with dicyclohexyl-lB-Crown- 13 6 are more stable in aqueous solution than those of uni- valent cations. However, comparing monovalent and bivalent cations having small but similar radii, the monovalent cation forms the more stable complex (i.e. Na+ > Ca2+) with dicyclohexyl-lB-Crown-G (6), because the divalent ion is much more solvated than the monovalent ion. b. Structure and Stiochiometry of Alkali Cation- Crown Ether Complexes Because of strong interaction between alkali metal cations and the cyclic polyether complexing agents it is often possible to prepare crystalline salts that contain the complexed cation. The crystal structures of a number of such salts have been determined (22-28). The crystal- line structure of macrocyclic polyether complexes can pro- vide useful information about the structure of the cationic complex, and its interaction with the anion and with sol- vent of crystallization as well as the stiochiometry of the complex. Evidently the stiochiometry of these complexes depends on the relative sizes of the cation and the macro- cyclic cavity, the flexibility of the crown ether molecule, and the nature of the anion and of the medium (29). How- ever the stiochiometry of the resulting metal-crown com- plexes is determined basically by the fit of metal cation in the cavity oftflmzligand. For those having the best fit, 1:1 (metal ion:1igand) complexes are considered to 14 be formed in solution as well as in the solid state. For a given size of metal cation, as the cavity size of the crown ether increases 1:1 complexes are formed, but the ligand increasingly tends to fold around the cation. Such folding of the ligand has been observed for dibenzo-24- Crown-8 in complexation with the Na+ ion (30). A similar behavior (twisted conformation) of the polyether chain in the solid potassium iodide complex of dibenzo-30-Crown-10 has been demonstrated by Bush andflrruter(3l). Live and 13C NMR studies for the Chan (32) carried out PMR and complexation of K+ with dibenzo-30-Crown-10 in different solvents. The conclude that the structure of (K-DB30C10)+ in solution is similar to that in the crystalline state in which the K+ ion is coordinated with all of the oxygen atoms of the ligand so as to form a "wrap around" complex. Figure 4 shows the crystalline structure of potassium iodide-dibenzo-30-Crown-10. A 2:1 (metal ion:ligand) complex (or bimetalic complex) may be formed when the size of the cation is much smaller than the cavity size of the ligand. As illustrated in Figure 5, in such a compound, two cations are complexed in the unfold cavity of the ligand. The bimetalic pro- ducts has been found by X-ray analysis of the complexes (KNCS)2 (DBZ4C8) (33) and (Na-O-nitrophenolate)2 (DB24C8) (34) and as expected for the complex of sodium with DBBOClO (30). 15 Figure 4. Crystalline Structure of (KI-DB30C10). (0:: 26):]. U L] \_/ Figure 5. Structure of DBZ4C8-(K+)2- 16 23 13 A recent Na and C NMR study (35) of complexa- tion between DB30C10 and the sodium ion in nitromethane and acetonitrile solutions showed that three kinds of sodium DBBOClO complexes with the respective stiochiometry of Na°(DB30C10), Naz-(DBBOC10), and Na3-(DB30C10)2 are pre- sent in solutions. When the size of metal cation exceeds the cavity of crown ether, sandwich complexes, in which the cation is located between two molecules of crown ether, can form in solution, Figure 6. The formation of such complexes can + Figure 6. Sandwich Structure of (B15C5)2-K Complex. be easily detected by the alkali metal NMR techniques especially when the 1:1 complex is strong. The formation of sandwich complexes of Cs+ with lB-Crown-6 and K+ with lS-Crown-S in different solvents has been reported by Mei (36) and Shih (37). The formation of cation-crown ether complexes with stiochiometry other than 1:1 has also been reported by Pedersen and Frensdorf (l, 8, 10, 29). 17 C. Sulfur-Oxygen Mixed Donor Macrocyclic Ligands Donor atoms such as sulfur and nitrogen have been used to replace oxygen in crown ether rings in order to vary the cation binding properties of these ligands (10, 38-40). Substitution of sulfur for oxygen atoms in crown compounds markedly affects their complexing properties. Alkali and alkaline earth metal cations give no evidence of complexation when all of the oxygen donors are replaced by sulfur atoms (19). However, complexation is expected with the ligands containing both sulfur and oxygen as the binding sites. Several methods have been used to prepare sulfur- oxygen mixed donor ligands (41-45). In general, a dimercaptan or sodium sulfide is reacted with an Oligo- etheylene glycol dichloride in a medium of an appropriate base. Macrocyclic polyether sulfides containing two to four sulfur atoms and two to four oxygen atoms have been found to form 1:1 complexes with alkali, alkaline earth, and silver ions (10, 41). Ultraviolet spectra of some of the sulfur crowns in solutions of methanol has been studied (41) and it has been found that these spectra are little affected by the addition of the salts of alkali and alkaline earth elements, suggesting that the inter- action between these ligands and the alkali and alkaline earth cations is not strong in methanol. Complexation of 18 1,10-dithia-18-Crown-6 with potassium and silver ions has been studied in methanolic solutions by Frensdorff (10) using potentiometric technique. It was found that the silver ion has a greater affinity for the ligand than potassium ion. In this case, Frensdorff postulated that the bonding between 1,10-dithia-18-Crown-6 and silver ion is not purely electrostatic. Silver ion can form both ionic bonds with oxygen atoms and covalent bonds with sulfur atoms, but the potassium ion may only form ionic bonds. Izatt (6) compared the thermodynamic stabilities of the K+, Tl+ and Ag+ complexes formed by 15-Crown-5 and its sulfur analog (Table 2). A marked enhancement of the Table 2. Log Kf values for reaction of 15C5 and thia-15C5 with some univalent cations in water (6) Metal cation lSCS thia-lSCS K+ 0.74 (a) + T1 1.23 0.80 Ag+ 0.94 5.2 a . No measurable reaction. . . . . + stability of the Ag+ complex relative to that of either K or Tl+ complexes is observed as sulfur is replaced for oxygen. 19 In general, macrocyclic polyethers containing sulfur atoms are poorer complexing agents for alkali cations than their oxygen analogs, but they are good complexing agents for silver ion. In complexing ability, Pedersen (41) attributed the differences between the macrocyclic poly- ethers and their sulfur analogs to the differences in the sizes and electronegativities of oxygen and sulfur atoms. Oxygen is a smaller atom than sulfur and the C-O-C bond angle is greater than the C-S-C bond angle; therefore, in the sulfur substituted ligand, the symmetrical distribu— tion of the negative charge around the "hole" of the poly- ethers is disturbed due to the large size of the sulfur atom. In addition, the electronegativity of the oxygen atom is higher than that of the sulfur atom which makes the C-8 bonds less ionic than the C—0 bonds. Crystalline structures of some of the cyclic poly- ethers containing sulfur atoms and their complexes have been determined by several researchers. Dally and co- workers (46) determined the crystal structure of 1,4,7- trithia-lZ—Crown-4 (Figure 7). As shown in this figure Figure 7. Crystalline Structure of 1,4,7-trithia-12- Crown-4. 20 the oxygen atom is directed toward the center or the ring but the sulfur atoms are directed away from the polyether ring. X-ray analyses for 1,10-dithia-18-Crown-6 and 1,4- dithia-lS-Crown-S also showed that the large sulfur atoms are directed away from the ring (46). This may be attributed to the differences in bond length of the C-0 and C-8 bonds and in the van der Waals radii for oxygen and sulfur atoms. It has been shown that the mercury atom in the HgCl2 complex with 1,4-dithia-18-Crown-6 and the paladium atom in the PdCl2 complex with 1,10-dithia-18-Crown-6 bind only with two sulfur atoms and lying outside the cavity (47, 48). Therefore, binding of sulfur containing crowns to H92+ and Pd2+ may not be of inclusion type which may account for no "macrocyclic effect" (see page 30) for these thia-substituted ligands (47). D. Complexing of Metal Cations by Macropolycyclic Ligands- Cryptands Soon after Pedersen's publication on the synthesis of the first crown ethers in 1967, Lehn and co-workers re- ported the synthesis of macrobicyclic ligands, or cryptands (2, 49). These compounds also form complexes (cryptate) with the same cations as the cyclic polyethers. In cryptates, the complexed cation is enclosed inside in a three dimen- sional cavity, whose size, shape, and the nature of binding sites determine the stability and the selectivity of the complexes. The size of the cavity can be varied by changing 21 the length of the ether bridges. The three bridges of a bicyclic ligand shield the cation from the medium if the cation fits or is smaller than the diameter of the cavity. The cryptand compound shown in Figure 8 is called cryptand-222 (K = m = n = 2) or C222 where the "C" refers to cryptand and "222" stands for the number of oxygen atoms in the chains. The cryptand-222,which contains six oxygen atoms in the bridges and two nitrogens at the bridge heads,is a bicyclic ligand, therefore it is called a [21-cryptand. Ligands containing three and four macro- cycle are called [3] and [41-cryptands respectively (50). C) n K = m = 0, n — 1 C211 (1.6 A°) K = m = l, n = 0 C221 (2.2 A°) K = m = n = 1 C222 (2.8 A°) Figure 8. Structure of Cryptands (with internal diameter). Complexation constants of cryptands have been de- termined mainly in water and methanol solutions. Lehn and co-workers (2, 51, 52) have studied the formation constants of alkali metal cryptates in aqueous and methanolic solutions 22 using potentiometric technique. Stability constants of com- plexes of a different cryptands with cesium ion in a number of solvents are reported by Hsu (53). In general, cryptands form much stronger complexes with the alkali ions than the crown ethers. Complexation studies of C222, C221, and C211 with the lithium ion were performed in water and several non- aqueous solvents by Cahen et a1. (14) using 7Li NMR tech- nique. Their results revealed that the chemical shift of the lithium ion complexed by C211 is solvent and anion in- dependent. Cryptand-le has a cavity size nearly equal to that of the unsolvated lithium ion. It is expected, there- fore, that the lithium ion is completely isolated by the ligand. On the other hand, with C222 and C221, whose radii are much larger than that of the lithium ion, the resonance frequency of the complexed lithium ion does de- pend somewhat on the solvent, suggesting that in these cases the lithium ion is not completely shielded by the ligand, and the loose structure of the Li+-C222 and Li+- C221 complexes permits the solvent molecules to approach to the metal ion so as to affect its resonance frequency. Popov and co-workers (14, 54-56) studied the com- plexation reaction between cryptands and alkali cations in several solvents using alkali metal NMR techniques. As with the crown ethers, they found that the stability con- stants of appropriate complexes are greater in the dipolar protophobic than in the protophilic solvents. 23 It has been suggested that the mechanism of com- plexation is complicated because of conformational equi- librium in the ligand (52, 57). As shown in Figure 9 both the free macrobicyclic ligands and their complexes may exist in three forms, exo-exo (or out-out), endo—endo (or in-in), and exo-endo (or out-in), differing by the orientation of the nitrogen bridgheads toward the inside or outside of the intramolecular cavity (58-60). Macrobicyclic ligands in their in-in conformation contain an internal cavity of about spherical shape (61). Therefore, they should be suitable for the spherical ions such as alkali and alkaline-earth cations. The exo-exo “by“ ”Q” “----i" exo-exo exo-endo endo-endo Figure 9. The Configurations of Cryptand C222. forms in which the lone pairs of the bridgehead nitrogens are directed away from the cavity would not be expected to complex appreciably with metal ions. X-ray analysis of the crystal structures of a number of cryptates showed that the cation is indeed located in the center of the molecular cavity of the macrobicycle in the "in-in" form (62-65). In addition to preparation of the macrobicyclic ligands, Lehn and co-workers also synthesized tricyclic 24 cryptands (50, 55, 57), and in several papers, have des- cribed the properties of these compounds. The cavity of the tricyclic ligands can be cylindrical or spherical as represented in Figure 10. The cylindrical ligands are formed by two macrocycle linked through two bridges, and contain three cavities, two on the top and the bottom, and the third in the center. The size of the cavities can be changed, by varying the size of the macrocycles and the length of bridges. “an. o (0th L. 4.3) 0:34;) NM. 6’10) {0 ”YSJ Léfij Figure 10. The Structures of (a) Cylindrical, and (b) Spherical Macrotricyclic Cryptands. As shown in Figure 11 these ligands can form mononuclear or binuclear complexes. In a binuclear cryptate the two macrocycles each serve as a receptor site for a cation. Values of log of stability constants for successive formation of mononuclear and binuclear [3]- cryptates of alkali and alkaline earth metal cations have been determined in methanol/water mixture by Lehn and Simon using ion selective electrodes (68). Their results 25 .Acnmqv ccmummuo HMOfluccflawo mo moxOHQEOO upmaoscocoz Ca mmmooum mmcmcoxm coflumo HaasomaofimuucH .Acsmav pcmummuu OflHOSOouomz HNOAHCCAHAO mo cumumwuu HMOHODCflm can Homaoscocoz mo coHuoEHOh m>wmmmoocm aux z:\/O\I/O\/\z .3. z 0 m u I m :... X ......n..m...n.... A128 F\ x I’\x .- 0.. (.6 I 0’. C C I. I o I o. a a no no 0. ,.... 0| ”’0'. ........ rec/koszu /\\ {CO/K .NH musmfim .HH musmflm \\)/O\]Jxx\l/ \\)/O\1J(R\l/ l\JO\l)d\/£2Jv .X 26 showed that the first formation constant of the inclusion complex of a given ion is somewhat larger than the second one. This indicates that the inclusion of the second ion is almost as easy as that of the first one. In mononuclear complexes of [3]-cryptates an intromolecular cation exchange may occur in which the cation exchanges between two sites. The cation exchange rate between two rings inside the cavity of a [3]- cryptate has been studied by 13 C NMR technique (68). The spectral changes showed an internal cation exchange be- tween binding sites on two rings inside the cavity of the ligand as represented in Figure 12. The spherical macrotricyclic ligands contain a single central cavity (Figure lO-b) and these ligands should be especially well suited for the alkali and alkaline” metal cations which are spherical in shape. Ligands con- taining spherical cavity were prepared by Graf and Lehn + , Cs+, NH4+ and Ba2+ ions. These compounds also form very stable complexes with (50). They complexed K+, Rb anions. The halide complexes are reported to be quite stable in water (50). E. Thermodynamics of Cation-Macrocyclic Interaction Thermodynamics in solution of macrocyclic polyether complexes has been extensively studied during the last ten years (3, 6, 10, 69). Most of the studies have been carried out in water, water/methanol or methanolic solutions 27 (47), consequently little is known at the present time about the thermodynamics of complexation in nonaqueous solvents. The equilibrium constant for the complexation of the Cs+ ion by C222 has been studied as a function of temperature in acetone, propylene carbonate, and N,N— 133Cs NMR techinque (54). dimethylformamide solutions using The AH: and ASE values, obtained in this work, showed that the complex is enthalpy stabilized but entrOpy destablized. Thermodynamics of complexation of Cs+ ion with large macrocyclic polyethers, dibenzo-BO-Crown-lo, dibenzo-24- Crown-8, and dibenzo-Zl-Crown-7 have been studied in five 133Cs NMR technique (70). It was nonaqueous solvents by found that the stability of the 1:1 complexes formed be- tween Cs+ ion and these ligands increases with decreasing temperature and in all cases negative AH; and A83 values characterize the formation of such complexes. It was con- cluded that as the cyclic polyether ring size increases, the entropy of complexation becomes more negative, suggest- ing that significant conformational changes of the ligand may be important in the formation of these complexes. An excellent interpretation of the enthalpic and entropic contributions for complexation in aqueous solu- tions has been discussed by Kauffmann and co-workers (71). The stability constant (Kf) of the complex is related to the net changes of free energy, AG°, enthalpy, AH°, and entropy, AS°, according to the well known equation: 28 -RT an = AG° = AH° - TAS°. From these functions important conclusions may be taken about the various factors governing complex formation, such as solvation effects, the character of the co- ordinate bond, and the changes of the structure often taking place during complex formation. Stability constants, from which the AG° values are calculated, provide a direct measure of the extent of complexing in solution. The measured enthalpy change for a cation-ligand reaction in solutions reflects (a) the bond energy of the cation-donor atom bonds, (b) solvation energy of the reactants and products, (C) the contribution arises from changes in ion-solvent interaction beyond the first solvation or ligation shell , (d) changes in interbinding sites repulsions. Several factors contribute to AS°, including (1) ligand and cation desolvation, (2) solvation of the complex, (3) translational entropy loss on formation of a single complex from two species, (4) changes in ligand internal entropy upon complexation caused by orientation and conformational changes. The cyclic polyethers form much more stable com- plexes than do their corresponding open chain analogs. Morf et a1. (72) and Izatt et a1. (47) have compared log K values for the reactions of some macrocyclic ligands + 2+ and their open chain analogs with K+, Na , Ba , and Cu2+ ions. As illustrated in Table 3 the stability constants 29 Table 3. Comparison of Stability Constants of Metal Cation Complexes of Macrocyclic Ligands with Their Open Chain Analogs (72,47) Log K Ligand Cation Solvent (at 25°C) (‘0 K+ MeOH (99 Wt. %) 2.27 0 ’1 H20 (1 Wt. %) Na+ " 1.0 I + 0"? Ba2 " 2.51 (‘0’\. K+ " 6.05 " + E 3 Na " 4.33 0 0 Ba2+ " 7 0 KIOd . 5W Cu2+ MeOH (80 Wt. %) 1.15 < H20 (20 Wt. %) \__/s_\ Cu2+ MeOH (80 Wt. %) 3.48 C H20 (20 Wt. %) c "”2 N < Cu2+ H 0 20.1 NH2 2 I \ . NH “N 2+ Cu H20 28.0 NH Ifli \.J /""\0 _Co 3W0, Ea2+ H20 4.80 (F\___/(’ ON Ea2+ H 0 9.50 N’\:°\/\o’\j" 2 Kzo\_/0 30 of open chain ligands are several orders of magnitude lower than those of their cyclic analogs. In order to emphasize the greater stability of macrocyclic complexes over the non-cyclic ones, the term "macrocyclic effect" was used by Cabbiness and Margerum (73). The macrocyclic effect is a Gibbs energy term, referring to the following reaction: (ML)n+ + L' -> (ML' )M + L non-cyclic macrocyclic macrocyclic non-cyclic and the enhanced stability of the macrocyclic complex may have its origins in both enthalpy and entropy terms (52). The macrocyclic effect has been explained in terms of entropy factors which oppose complete envelopment of cations by acyclic polyethers (73). The linear polyether does not envelope the cation as completely as the cyclic one because of electrostatic repulsion between its terminal groups and the unfavorable entropy change involved in forcing the linear into an almost cyclic structure. Al- though this effect was originally explained in terms of entropy factors which oppose complete envelopment of cations by acyclic polyethers, however, studies on macro- cyclic tetramines (74, 75) and tetrasulfide systems (76) revealed that the thermodynamic parameter (AH), coupled with ligand solvation effects, often play a greater role in the origin of the macrocyclic effect. 31 The enhancement of the stability of cryptate com- plexes over the crown complexes has been observed by Lehn and co-workers (21). The increase in complex stability of bicyclic ligand over monocyclic ligands which is even larger than the macrocyclic effect is called "cryptate effect" or "macrobicyclic effect" (21). This effect is also illustrated in Table 3. F. Nuclear Magnetic Resonance a. Introduction Solutions of alkali metal cation complexes have been studied by various experimental techniques. Potentiometric measurements have been done by Frensdorff (10) and Lehn and co-workers (2) for determination of formation constants of alkali ion complexes with many macrocyclic ligands. In recent years, however, nuclear magnetic resonance spectroscopy, as well as infrared and Raman spectroscopy have become powerful techniques for the study of the solva- tion phenomena and of complexation reaction in both aqueous and nonaqueous media (16, 77-85). Since nuclear magnetic resonance measurements are Specific for each nucleus, they can be used for quantita- tive and qualitative determinations of the species in solutions. Proton, 13C, and alkali metal NMR studies have [provided much detailed information about the nature of cation complexation by crowns and cryptands. Proton and 13NMR techniques can be used for ellucidation of structure 32 in solution as demonstrated, for example, by the work of Live and Chan (32) who were able to show the solution structures of complexes of some of the crown ethers. Nuclear magnetic resonance of the alkali cations and the thallium ion have been used extensively for study- ing ionic solvation, ionic association, preferential solva- tion of alkali cations and thallium ion and in complex formation with a variety of ligands, particularly macro- cyclic ligands in aqueous as well as nonaqueous solvents. A review of the alkali metal NMR has been published by Popov (77). An older review of the use of the NMR for the study of ions in different solvents has been published by Amis and Hinton (86). b. Chemical shift measurements of alkali metal nuclei Since the alkali metal chemical shifts are very sensitive to the immediate environment of the correspond- ing metal ion in solutions, they have been used for the studies of ion-ion, ion-solvent and ion-ligand interactions (77). For nuclei with an adequate NMR sensitivity, ob- servation of signals at low enough concentrations to allow determination of stability constants should be possible as demonstrated by Dewitte et al. for 133Cs (87). All of the alkali metals possess at least one 7 23 39 isotope with a magnetic nucleus, i.e., Li, Na, K, 87Rb, and 133Cs. The nuclear properties of the alkali elements are given in Table 4. In all cases the spin is 33 aso.em aom.a omo.aa eom.~ mom.ma mam.m~ ma.os ooH a.~a mo.ma cos sm.~a maa.o muoaxes.e eaa.o euoaxmo.m m-osxsm.a aa~.o ~HG.H meom.m sass.m aoam.o seam.m oom~.m ~\H Nxa «\m m\m «\m ~\m Hemom mumma hmsm Ram 62mm ass Amaze caoam mhsmmoaax ao.ea um mocmnvmuw OUCMCOmOm Rwy OOCCCCSEM Honduwz Camfim accumcoo um ma Ou >uw>fluamcom m>wumem Auzpv\£m .coumcmmfi Homacsc msu mo mmamfluasfi ch : ucmEOE oeumcmmz As~\c mo human Gav :flmm Hmmaosz .HOHODZ EDHHHCSB can Hamxad mo mmfluummomm meaosz .v:manma 34 greater than % and, therefore, the nuclei have a quadrupole moment and consequently, the resonance lines cannot be narrow. In practice, however, due to the extremely rapid and generally random molecular motions which averages the local magnetic and electric fields to very small values, 87 . Rb, the resonance llnes are narrow and, in the case of 7L1 and 133Cs the natural linewidth with the exception of are less than 1 Hz (16). Therefore in most cases the chemical shifts can be measured precisely. The position of a nuclear magnetic resonance signal is determined by the total shiedling that the nucleus under investigation receives from various sources. This shielding is expressed by the screening constant, 0. A general formula for O has been developed by Ramsey (88). According to Ramsey's equation the screening constant is the sum of the various diamagnetic and paramagnetic terms: 0 = Op + Ca. Saika and Slichter (89) divided the screening con- stant into three independent contributions represented by: The O term (shielding factor) is related to the local d diamagnetic currents in the molecule and arises from the induced magnetic field due to the circulation of electrons .around the nucleus; Up is the paramagnetic term (de- shielding factor) and is usually the dominant term, and 35 it arises from the interaction of ground state with excited electronic states in the presence of a magnetic field. Contributions from other atoms to the shielding of the resonant nucleus are contained in 00. Unless one is considering proton shifts, O can be ignored. For 0 the heavier nuclei, Up is so much larger than Od that the later term can be ignored (90). Kondo and Yamashita (91) suggested that the para- magnetic shift term is caused by the repulsive forces be- tween the Closed shell of the ions. These forces act over a short range and compete with electrostatic attractions which tend to reduce the separation of ions containing opposite charges. The paramagnetic term for an alkali nucleus can be written according to the following equation: o = -16 62 (AEI'l s2 P . HP HP 92 in which a is a constant = 2 where m and e are 2mC the mass and charge of the electron respectively, np is the expectation value for the outermost P-electron of the element, (AB);; 'is the average mean excitation energy from the np to the (n+l)p orbital, and S is the overlap integral between P orbital of the alkali ion and the neighboring ions. Deverell and Richards (92) who modified the theory of Kondo and Yamashita, provided a qualitative interpreta- tion of cation chemical shift in solutions. They suggested that the chemical shift at concentration c, relative to 36 the free ion can be written as: o — 16 a np (AB)np [Z Ai_j + (Ai_H 0 Ai_H 0)] 3 2 2 where AC refers to the appropriate sum of the overlap i-j integrals for the orbital of the central ion (1) and all other ions present in solution which may have the same charge or the Opposite charge of the central ion. The c _ o . . . Ai-HZO Ai-HZO indicates changes in the effect of solvent-ion overlap integrals. Since (r13>np and (AB);; both increase with increasing atomic number (93), the magnitude, and therefore the relative importance of . . . .+ . +. . Op increases in gOIng from L1 ion to Cs ion. Since the range of the chemical shifts varies directly with the atomic number of the alkali nucleus, the NMR measure— ments are particularly sensitive in the case of cesium— 133 nucleus. In addition the narrowness of the resonance line permits very accurate determinations of the chemical shifts. C. Thallium-ZOS NMR Measurement The thallium ion has been proposed as a probe of the role of the alkali ions in biological systems (94) because its chemistry is analogous to that of the K+ ion (95). The ionic radii of K+ and Tl+ ions are similar (1.33 and 1.40 A° respectively) (96), and the chemical preperties of these two ions are sufficiently alike that + + . . Tl can replace K in several enzymes Without loss of 37 activity (96). The NMR properties of 205T1 are given in Table 4. Thallium-205 is an ideal NMR probe for studies of ionic solvation, ionic association, and complex forma- tion in aqueous as well as nonaqueous solvents, because the change in its resonance frequency is very sensitive to its chemical surrounding, and it has a Spin = %, high natural abundance, and high sensitivity. The solvent de- pendent chemical shift of 205T1 is much larger than alkali nuclei. For example the change in its resonance frequency in going from water to pyridine is approximately 782 ppm. By comparison, with the change in chemical shift of 23Na in these two solvents which is about 1.26 ppm (81). There- fore, the greater sensitivity of the thallous chemical shift to the environment of the ion makes it a better system than alkali metal ions as regards of subtle changes in its environment. Chemical shift measurements of 205T1 have been made by Freeman et a1. (97, 98) for different thallium salts in aqueous solutions. Bacon and Reeves ( 99) who studied the spin-lattice 205T relaxation of 1, reported that the spin-lattice re- laxation rate and the spin-spin relaxation rate (R2) of the thallous ion are very sensitive to the disolved oxygen in aqueous solutions. It was found, however, that the relaxation rates are independent of the anion, the con- centration, and isotope substitution in the solvent (i.e., 203 1320 for H20) or in the resonant nuclei (i.e., T1 for 1205T1). In addition,chemica1 shift measurements have 38 been made and the effects of the addition of complexing agents (ethylendiamine and 0-phenanthroline) on the chemical shift and relaxation of thallous ion in aqueous solution have been discussed by Chan and Reeves (100). Zink et a1. (101-103) studied preferential solva- tion and the solvent dependence of the thallium-205 using 205T1 NMR technique. They found that the solvent de- pendence chemical shift of the thallous ion is over 2600 ppm depending on the particular thallium salt and solvent under study, and it correlates linearly with the relative solvating ability of the solvents for T1+ ion. Solvation of the thallous ion hidilute solutions of several binary solvent systems was studied by Briggs and Hinton (84, 85) by 205T1 NMR spectrOSCOpy. An attempt was made to separate solvation effects related to the Lewis basicity of the solvents from the structural changes in the solvation Sphere. It was found that T1+ ion is preferentially solvated by DMSO rather than pyridine, despite the latter's greater electron-donating ability. This was explained in terms of structural effect or strict effect related to the size of pyridine molecule. Dimethylsulfoxide is known to be associated in the pure liquid due to dipolar interactions through S-O bond (104). Introduction of pyridine into neat ‘DMBO results in the break up of the polymeric structure of the latter via a dipole interaction, resulting in an enhancement of donicity of DMSO in the mixture (104). In 39 addition, pyridine is more structured in mixed solvent system than in the pure solutions, and thus less able to solvate the ions when mixed with other solvents (85). Thallium-205 NMR method has been used by several researchers for studies of the complexes of T1+ ion with antibiotic ionophores such as, nonactin, monactin, dinactin (105) and valinomycin (96). Srivanavit and co-workers (106) determined the stability constants of thallium complexes relative to formation constants of other uni- valent cations complexes for a number of macrocyclic crown ethers in various solvents using NMR techniques. The effect of the solvents on the binding constants was dis- cussed in terms of solvation of the ions and the ligand. CHAPTER II EXPERIMENTAL PART 40 A. Synthesis and Purification of Ligands a. Synthesis of 1,4,7-trithia-12-Cr9wn-4 (TT12C4) The complexing agent 1,4,7-trithia-12-Crown-4 was prepared in a manner similar to that reported by Bradshaw and co-workers (42). This material was synthesized from Bis(2-chloroethyl) ether (Aldrich) and Bis(2-mercaptoethyl) sulfide (K and K) as illustrated in the following reaction C1CH2-CH2-O-CH2-CH2C1 + HS-CHZ-CHZ-S-CHZ-CHz-SH S S gig—1%?» 2NaCl + 21420 + I: D 2 5 C) S U A solution of sodium hydroxide was prepared by dissolving 12.5 gram of sodium hydroxide (Dark Brothers) in 400 ml of ethanol and was placed in a large three- necked round bottom flask. An equimolar mixture of the Bis(2-chloroethyl) ether (0.11 mole, 15.75 gram) and Bis(2-mercaptoethyl) sulfur‘ (0.11 mole, 17.1 gram) was dissolved in 300 ml of ethanol and added drOpwise to the NaOH solution over a long period of time (about 6 hours). The reaction flask was equipped with a mechanical stirrer as well as with a nitrogen gas inlet and outlet. In order to increase the yield of the product the reaction was carried out at high dilution with vigorous stirring, under a nitrogen atmOSPhere. Once the two solutions were 41 42 completely mixed, the reaction mixture was refluxed for about 8 hours and then was allowed to cool to room tem- perature. The excess base was neutralized with hydro- chloric acid, and then the resulting mixture was filtered. The solvent was removed using the Rotovap-R Bfichi. The residual viscous liquid was extracted with three 200 m1 of diethyl ether. The combined ether extracts were dried over anhydrous magnesium sulfate and then filtered. After removal of the ether and recrystallization of the remaining yellow solid in benzene, the pure product was obtained. Following the purification by recrystallization, the ligand was dried under vacuum over barium oxide (Fisher Scientific Co.) for 3 days at room temperature. Its melting point (measured on the Fisher Johns Melting Point apparatus) was found to be 91°C which is identical to the literature value (42). A 0.1M solution of TT12C4 in deuterated chloroform (Aldrich) was used and the chemical shifts for three kinds of hydrogen (a, b, and c, Figure l) were measured relative to TMS (Merck Co.) as an internal standard. The NMR assignments were exactly the same as reported by Bradshaw (42). b. Purification of commercial ligands Eighteen-Crown-six (18C6) was obtained from the Parish Chemical Company. The ligand was purified by form- ing a complex with acetonitrile (Matheson, Coleman and Bell) (107). The fine white crystals of the 18C6.MeCN 43 complex were dried under vacuum for 48 hours to remove the weakly bound MeCN from the complex. The melting point of the purified product was 38-39°C in satisfactory agreement with the reported value of 39°C (107). Dithia-lB-Crown-G was obtained from Parish Chemical Company and required vacuum drying for two days. The complexing agents, dibenzo-24-Crown-8, and dibenzo-21-Crown-7 were received from Parish Chemical Company and they were purified by recrystallization from normal heptane (Mallinckrodt) and then dried under vacuum over barium oxide (Fisher Scientific Co.) for three days. The melting points of the fine white crystals were found to be 114° and 106°C, respectively, which are the same as the literature values (29). Cryptand—222 (E.M. Laboratories, Inc.) and Cyclo— (tetraethylen-glycol-Z,6-Pyridinedicarboxylate) (Parish Chemical Company) were purified by two recrystallizations from n-hexane (Fisher Scientific Co.) and then vacuum dried for several days. Dibenzo-27-Crown-9 (Parish Cehmical Company) was recrystallized from reagent grade n-heptane and dried under vacuum for several days. 8-92129. Lithium perchlorate (Fisher Scientific Co.) was oven dried at 190°C for several days. Anhydrous silver nitrate (Baker, AR) was dried over P205 in a vacuum oven at 65°C for one day. Thallium (I) perchlorate (K & K) after purification by recrystallization from deionized 44 distilled water was dried at 110°C for twenty four hours. Sodium tetraphenylborate (T.J. Baker) was dried in a vacuum oven at 60°C for three days. Cesium thiocyanate (Rocky Mountain Research, Inc.) was recrystallized from methanol and dried under vacuum at 50°C for 48 hours. After drying, the salts were stored in vacuum desicators charged with granulated barium oxide (Fisher Scientific Co.). C. Solvents The solvents were purified and dried according to the following procedures: Propylene carbonate (Aldrich), nitromethane (Aldrich), dimethylformamide (Mallinckrodt), dimethylsulfoxide (Fisher), and Pyridine (Fisher) were re— fluxed over calcium hydride (Fisher) under reduced pressure over night and then fractionally distilled and further dried over freshly activated molecular sieves (Lind type 4A). Molecular sieves were washed with distilled water, then oven dried, and finally activated to 500°C under a flow of dry nitrogen. Acetone (Fisher) was refluxed over calcium sulfate (W.A. Hammond Drierite), fractionally dis- tilled, and dried over activated molecular sieves. Tetrahydrofuran (Baker, analyzed reagent) and acetonitrile (Mallinckrodt) was refluxed over calcium hydride, frac- tionally distilled and further dried over activated molecular sieves. Tetramethylguanidine (Eastman) was re- fluxed over granulated barium oxide (Fisher) for about 30 45 hours and purified by fractional distillation under reduced pressure. Methanol (Fisher or Mallinckrodt) was refluxed over magnesium turnings and iodine for ~ 24 hours and then distilled under nitrogen atmosphere. The distillate was allowed to stand over activated molecular sieves for ~ 20 hours. All of these solvents were stored in brown glass bottles and transferred in a dry box under a nitrogen atmosphere. Water contents and purity of the solvents were tested either by Karl Fischer titration using an automatic Karl Fischer Aquatest (Photovolt Corp.) titrator or by using a Varian Aerograph Model 490 gas chromatograph with a Porapak Q Column. The water content of the purified sol- vents was below 100 ppm in all cases. Acetonitrile-d3 (Stohler Isotope Chemicals) and dimethylsulfoxide-d6 (Stohler Isotope Chemicals) were used as received. D. Instrumental Measurements Alkali metal and thallium NMR measurements were performed on a highly modified Varian Associate DA-60 spectrometer in the Fourier Transform mode, operating at 23.31, 15.87, 7.87 and 34.61 MHz for 7Li, 23Na, 133Cs 20511 respectively. The instrument has a magnetic and field of 14 Kilogauss (1.4 Tesla), and it is equipped with a wide-band probe capable of multinuclear Operation (108). An external proton field lock was employed to 46 maintain the field stability. The NMR spectrometer is interfaced to a Nicolet 1083 Computer for data aquisition. Data treatment was done using the Nicolet FT-NMRD program (109). Non spinning 10 mm OD NMR tubes (Wilmad) were used. In the case of studies of temperature dependences 133Cs chemical shifts the temperature was monitored of the using a thermocouple housed in alO mm NMR tube and inserted in the solvent. During measurements a constant tempera- ture was monitored within : 1°C. Because the resonance frequency of the reference varies with the temperature an insulated reference tube (110) was used. In order to reach the equilibrium temperature, each sample tube was left in the probe for ~ 15 minutes before any measurement. Chemical shifts obtained for the different metal nuclei were referenced to 0.31 M aqueous solutions of TlNO for 205T1, 3.0 M LiClO for 7L1, 0.5 M CsBr for 3 4 133Cs, and 3.0 M NaCl for 23Na NMR measurements. However, all the Chemical shifts for sodium—23, Cesium-133, and thallium-205 reported in this thesis are ultimately re- ferenced to infinitely dilute aqueous solutions of the Na+, Cs+ and Tl+ ions. All reported data are also corrected for differences in the bulk volume diamagnetic suscepti- bility of the solvent (111, 112). The following equation ‘was used to make these corrections. _ Zn r _ s acorr - 6obs + 3— (Xv Xv) (1) ‘where 6 and 6 are the corrected and observed corr obs 47 ooo.oI o~>.OI umumz mmm.oI 0mm.0I AOZBV wcwoflcmomahsuoEmuuoe mmm.0I mam.0I Away mewcwumm omH.0I cmw.oI Aumv mumconumo cmammoum mmo . OI Ham . CI A273 mcmzumfiouuwz mmv.oI mam.o| Amoozv Hocmsumz HvN.0I mom.0I AOmzov OCAxOMHSmamnumEflo mom.OI oom.OI Amzdv mUflEmEHOMH>£quflQ omm.OI «mm.0I Azaom ossHo> stm muco>aom WSOwum> How ooldo co cowuowuuou zuwaflnflummomsm Ofluwcmmefla .m manna 48 chemical shifts, respectively, and x: and x: are the bulk volume diamagnetic susceptibilities of the reference (aqueous) and sample (nonaqueous) solvents. Since the concentration of added salt was low in all cases, no correction for the contribution of the salt to the magnetic susceptibility of the solution was applied in these studies. The magnitude of the correction for various solvents is shown in Table 5. For the mixed solvents, the volume diamagnetic susceptibility of a given mixture was calculated by the following equation (113) V V ’531"=\7—-3“v*"<3+v*—5—v“'x3 <2) A B A B where xmin is the calculated volume susceptibility of the mixture, VA and VB are the volumes of solvents A and B respectively, XA and XB are the volume diamagnetic susceptibilities of the pure solvents A and B respectively. Downfield (paramagnetic) chemical shifts from the reference are designated as positive. The PMR spectra were obtained at 25°C on a Varian T-60 spectrometer. Spinning 5 mm NMR tubes (Wilmad) were used. Chemical shifts were measured relative to TMS (Merck Co.) as internal standard. The shifts reported in this ‘thesis are given in ppm at a magnetic field of 14.09 KG. 49 E. Sample Preparation All samples were weighed out into volumetric flasks using an analytical balance. In view of the hydroscopicity of the nonaqueous solvents and of the rea- gents, all of the solutions were prepared in a dry box under a nitrogen atmOSphere. The mixed solvents were prepared by taking a certain volume of a solvent and mixing with the second solvent to give the desired volume per- cent of the binary solvent. Solutions containing both salt and ligand were usually prepared (if the ligand was soluble enough) by mixing apprOpriate columns of stock solutions followed by dilution to the mark with pure solvent or solvent mixture. When the ligands were only sparsely soluble, the samples were prepared by weighing out the various amounts of the complexing ligand into a 2 ml volumetric flask followed by dilution with the metal ion solution. After dissolution of the ligand, the solutions were transferred to 10 mm NMR tubes, capped, and wrapped with teflon tape to prevent both contamination by atmospheric water and solvent evaporation. In the case of studies of competitive solvation + O O O 0 of Cs ion in water-nonaqueous mixed solvents, the solutions were prepared by weighing a snap cap vial, adding the de- sired volume of one solvent, weighing, then adding the desired volume of another solvent and weighing again. Knowing these weights, the number of moles of each sol- vent and the resultant mole fraction were calculated. The 50 appropriate amount of cesium salts were weighed out into a 2 m1 volumetric flask and the desired solvent mixture was added to the mark. F. Data Handling Chemical shift data obtained from complexation studies were analyzed on a CDC-6500 computer using a non— linear curve fitting program, KINFIT (114) to obtain the formation constant of complexes. A linear least squares program was used to obtain the enthalpies and entropies of complexation. CHAPTER I I I MULTINUCLEAR MAGNETIC RESONANCE STUDY OF COMPLEXATION OF Na+, Cs+, Tl+, Li+ AND Ag+ IONS BY THIA-CROWN AND CROWN ETHERS IN NONAQUEOUS SOLVENTS 51 A. Introduction Since the synthesis of macrocyclic polyethers by Pedersen (l), the properties of these ligands and their complexes have been under active investigation because of the possible chemical and biological applications of their unusual ion complexation and transport effects. For these applications one needs the knowledge of the effect of sol- vent, cation, anion, and ligand parameters on the complexa— tion reactions between the metal ions and macrocyclic ligands. It has been found (14, 56, 115-117) that nuclear magnetic resonance of alkali nuclei, such as 7Li, 23Na, and 133Cs, is one of the most powerful techniques for the complexation studies of corresponding ions with macrocyclic ligands in nonaqueous as well as in aqueous solutions. Thallium-205 NMR has also become an important probe for investigation of solvation phenomena, solution structure, and complexation of the thallous ion. The thallium-205 chemical shift is very solvent dependent and it is very sensitive to small changes in the chemical environment of the Tl+ ion. Both the chemical shift and the line width of the nuclear resonances of alkali metal ion nuclei and 205Tl+ ion give useful information about the ion-ligand, ion-solvent, and ion-ion interactions. The work presented here deals with the use of alkali metal nuclei, thallium-205, and proton NMR techniques 52 53 which not only provide the binding constants of the dif- ferent complexes but they also give useful information on the effects of the solvent, the cation type, the type of the donor atoms, and the ring size of the macrocyclic ligands. In this chapter we will compare the ability of DT18C6 and 18C6, to form complexes with thallium, sodium and cesium ions in several nonaqueous solvents. We also describe the complexation strength of TT12C4 with silver and lithium ions in nonaqueous solvents. A comparison of the ion-binding properties of the macrocycles TT12C4 and 12C4 to lithium ion in a wide variety of nonaqueous solvents will be made. B. Results and Discussion l. Complexation of Na+ ion by DT18C6 The sodium-23 chemical shift-mole ratio data are given in Table 6. The variation of the sodium-23 chemical shift as a function of DT18C6/Na+ in various nonaqueous solvents is shown in Figures 13 and 14. It is obvious that the solvent plays an important role in the complexation proCess. For example, except for DMSO solutions, the sodium ion resonance shifts upfield or downfield (depending on the solvent) with the increasing concentration of the macrocycle until a certain mole ratio of Ligand/Na+ is reached. The more pronounced curvature in the mole ratio plot in NM solution (Figure 13) compared to those of other solvents is an indication of a stronger 54 SHNo.o u vemmz mo coflumnucmocouo Smoo.o n vemmz mo coflumuucwocoum Smmo.o n vemmz mo COHumnucmOCOOQ 2mo.o n vemmz mo coflumnpchCOOU .zmo.o u eemmz mo :Oaumupsmocoum he.hI Hm.m mm.v| m~.m Hm.nl wv.m mm.hI oa.m mo.v I ma.v no.mI mo.v mm.OI mm.m nm.vI mv.~ vs.>| mo.~ hm.>I Hm.m oa.e I mo.m mo.wI No.m hm.0I ma.m vm.v| Hm.a em.nI nm.a hm.hI mo.m wa.v I wo.m mo.mI vm.m mm.OI mm.H Hm.vI om.H Ho.mI nm.a Hw.nl hv.H mv.v I mm.a ma.mI mm.H Ho.o mH.H vw.¢I om.a mo.mI mo.H om.hI mm.a em.q I 5v.H Hm.mI vm.a mm.OI om.o m>.vI Ho.H Hm.mI mh.o mo.wI mo.a o>.e I e~.H mm.wI o~.H mH.OI hh.o o>.eI >>.o mH.mI mm.o mm.>I mm.o mm.m I mm.o om.wI ~o.a vm.oI mm.o on.eI Ho.o mm.mI he.o Nn.hI om.o mm.m I m>.o hh.mI om.o do.OI mm.o Ho.vI mm.o o>.mI em.o hh.>I Hm.o om.m I mm.o mm.mI om.o mo.OI oo.o Hm.vI oo.o oa.mI oo.o mm.hI oeso vo.vHI oo.o mm.>I oo.o agave +m2\q Asmmvw CZ\A AEmmvo +MZ\A Afimevc MZ\A Aammvm CZ\A Aammvo +mZ\A womzo mom—)5 O Q Q22 m muCOPHOm WDOHHC> CH mOmHBQ mo mocmmmum map as mumuonahcwbmmuume Eofipom MOM puma unflcm HmofiEmnu I Owumm 0H0: .o OHQCB 55 l5.0 ‘ l2.0 A 9.0 PC E 60— AN CL CL . " A ('0 + NM 3.0 ~— -3.0 I l I J 0.0 l.0 2.0 3.0 4.0 [DTI8C6]/[NO+] Mole Ratio Figure 13. Chemical shifts of sodium-23 as a function of DT18C6/Na+ mole ratio in various solvents. 56 8.0 (I AC 7.2-— -8 (ppm) 4.2 —- J J l 0.0 LO 2 .0 3.0 4.0 [DTI8C6] /[No+] Mole Ratio Figure 14. Sodium-23 chemical shifts vs. DT18C6/Na+ mole ratio in AC and DMF. 57 interaction between Na+ and DT18C6 in NM than in all other solvents. As shown in Figure 13 the frequency of the 23Na resonance in DMSO is essentially independent of the DT18C6/Na+ mole ratio. Therefore, there is no evidence for the formation of a complex in this solvent. The variation of the chemical shift with Ligand/ Na+ mole ratio was used to obtain the formation constants of (DT18C6-Na)+ complex. Assuming that only cation-ligand interactions are important and the rate of exchange of the metal ion between the two sites (free cation in the bulk solution and the complexed cation) is fast to the NMR time scale, the observed chemical shift is related to those of the free cation and the complexed cation by the following equation 6obs = 6MXM + 6MLXML (l) where sobs is the observed chemical shift, XM and XML are the fractions of the cation in the free and complex states and 6M and 6ML are the chemical shifts characteristic of the two environments. Given that XM = CM/C; the following expression can be derived 6 = SE (0 - 0 ) + 0 (2) obs C; M ML ML where CM and c; are the concentrations of free metal ion and total metal ion. The stability constant is de- fined as the equilibrium constant (in lit/mole) for the following reaction W + L—e.._ m.“ (3) 58 where M+ and L represent uncomplexed cation and ligand, and ML+ is a 1:1 (metal ion : ligand) complex. The concentration equilibrium constant is given by the usual expression [ML+ CML Kf=—T__]'=E'E— ‘4) [M JIIJ .M L where CML’ CM, and CL are the molar equilibrium con- centrations oftfluecomplex, free cation, and free ligand respectively. By a simple algebraic substitution, the above expression may be rewritten in the form of t CM ' CM Kf = t t (5) CM(CL - CM + CM) where CE is the total concentration of the ligand Using the mass balance equations, we can derive equation (6) t t 2 t2 2 t2 2 t t debs = [(KfCM - Kf CL - 1) i (KfCL + KfCM - 2KfCMCL + 6 - 6 t t a M ML 2KfCL + 2KfCM + l) ](;“"f‘—’) + 6ML (6) KfCM which relates the observed chemical shift to the formation constant (Kf), the total metal cationtflmeligand concentra— tion (CE and CE respectively). the chemical shift of the free cation (0M), and the chemical shift of the complexed . . t t cation (GML). Since Gobs' CM, CL are known and the value of 0 can be measured, using the metal salt solution M with no ligand, equation (6) contains two unknown parameters, 59 5ML and Kf. A nonlinear least-squares program KINFIT is used to solve the expression by using the experimental t t . . obs' M’ CL and 6M and adjusting Kf and 6ML values until the calculated chemical shifts correspond to the experimental values within the error parameters, 6 C limits of the measurement. The results of the above cal- culations for (DT18C6¢Na)+ system in various solvents are given in Table 9. The data collected in Table 9 show that the stability constants of (DT18C6.Na)+ complex among various solvents decrease in the order NM > AC > PC > AN > DMF > DMSO. This behavior reflects the much stronger cation solvation by DMSO and DMF, compared to other solvents, with which the polyether has to compete. Various scales of the solvating ability of solvents have been suggested. A quantitative measure of the solvat- ing power of solvents is provided by the so-called "donor number", D.N, which seems to agree quite well with the behaviour of alkali cation complexes in nonaqueous solu- tions. The donor number which has been proposed by Gutmann (118)is defined as the negative enthalpy value (in Kocal/Mole) for the 1:1 adduct formation between antimony pentachloride and the solvent molecules in 1,2-dichloroethane as an inert solvent 5 + SbCls—hflg’a S.SbC15 -AH (K cal/Mole) 5 Donor Number S.SbC1 60 The Gutman donor numbers and the dielectric con- stants as well as the structural formula of the non- aqueous solvents are shown in Table 7. Generally, we expect that in solvents with high donor ability and di- electric constant the stability constant of the complex should decrease due to the competition between the ligand and the solvent molecules for the metal ion. The results obtained in this work fit this generality. As illustrated in Table 9 the resonance frequency of the complexed 23Na+ ion is solvent dependent which in— dicates that the metal ion is only partially insulated from the solvent and the solvent molecules can still interact with the vacant coordination sites of the com- plexed cation. 2. Complexation of Cs+ ion by DT18C6 The cesium-133 chemical shift-mole ratio data for complexation of cesium ion by DT18C6 in acetone, propylene carbonate, nitromethane, and dimethylsulfoxide are shown in Table 8. Figure 15 shows the mole ratio plots for the (DT18C6.Cs)+ complex in various solvents. In all cases the plots show little curvature with no observable break at any mole ratio, which in- dicates the existence of a very weak interaction between the cesium ion and the ligand. In the case of acetone, propylene carbonate and nitromethane solutions, the plots of chemical shifts versus ligand to metal ion mole ratios 61 o.m¢ m.mm motiw.lo : o Aomzov mofix0masmasnumafla mmo H.mm o.om m “Uu2nlw.lm Amzov mUHEmEHomawcumEHo mu o m n.~m s.m~ mo.::.o m Amomzv Hocmsumz .m n.0m o.sH mmoll.w.xuc.m Romy wcoumua O m m0 // N m0 mo ‘ W Amumcooumo otl. \\ owaomo Hoflpcmmoumlm.av o.mw H.ma 0 Away mumconumu mamammoum a o m.sm H.vH zouomm Azav mafluuflcoumoa m.mm n.m NOZIUM: fizzy mcmnumeouuflz unnumcoo m uwnEsz oauuomamflo H%c o :cmEusm musuosuum unm>aom mucm>aom cflmuuwo mo ucmumcoo Uflupooamflo pom .HOQ55z uocoo acmEuDo .musuosuum .h manna 62 . mocoum m MHH m mfinv .mHH mocmummmm Amv m.m> mm 2 3vO m noun: m2 m = m o.HH - moi/21.0.uznnu m Anzac mcfloacmsmamsumsmuume m \\ m mu U m z m.NH H.mm F: _ Away wcflpfluhm unnumcou AmvaQEDZ Ufluuomamflo Hosoo somEuso musuosuum ucm>aom 11 Acmscflucoov a magma Smoo.o uHmm mo coflumuucmocouo 63 Svao.o n uHmm mo coflumnucmocoun Emmo.o n pawn mo coflumuucoocoum on.m~I Ho.m oa.omI mH.~ om.m I hm.m mm.HvI on.m hm.omI >>.H ma.h I mo.m mm.wm om.m mm.¢vI am.m Hm.HmI mm.H mm.> I nm.~ oo.nm om.m v~.va mm.m om.~mI mo.H na.m I mo.m mm.mo mm.H hm.va mm.H om.mmI Hm.o mo.oaI hm.a om.mw mm.a v>.va mo.H mo.NMI m>.o vo.oHI mw.a hm.mm vH.H Hm.HmI hH.H om.MMI Ho.o hm.HHI Ho.a hm.mo mo.H oo.mmI mm.o mm.mmI mm.o om.HHI mm.o hm.mm mm.o av.va vw.c Hm.mmI om.o h~.mHI «m.o mo.hw vm.o hm.mmI mm.o hm.mmI mm.o om.mHI m~.o hm.mw oo.o hm.hmI oo.o vv.va oo.o mh.maI oo.o AEmmvm +m0\q AEamvc +mU\A AEmmvw +mU\A Aammvc +mU\q 00min 022 now MU¢ mucm>aom msomsqmcoz msowum> ca mowaao mo mocwmmnm on» :w zummo new name umwnm HMUflEmno I ofiumm mac: .m manna 64 Figure 15. Cesium-133 chemical shifts vs. DT18C6/Cs+ mole ratio in various solvents. -8(ppm) 600 550— 500— 450— NM 400— 350 330 $0 290 210— _ AC 1 I l l IO 20 3O 40 [DTI8C6] /[Cs*] Mole Reno 65 (Figure 15) show that complexation of Cs+ ion by the ligand yielded paramagnetic shift, resulting from an increase in the electron density around the cesium ion. In the case of dimethylsulfoxide, however, which is a solvent with high donor ability (DN = 29.8), no chemical shift variation results as the Ligand/Cs+ mole ratio is increased (see Table 8). The stability constants and the limiting chemical shifts for (DT18C6.Cs)+ are listed in Table 10. Since the cesium ion is rather weakly solvated because of its low charge density, the stability of the complex is only marginally dependent on the nature of the solvent. The large difference between the limiting chemical shift of the complex is a good indication that the complexed cation remains exposed to the solvent molecules. 3. Complexation of T1+ ion by DT18C6 The complexation between thallous ion and 1,10- dithia-lB-Crown-6 was studied in nitromethane, acetone, acetonitrile, dimethylformamide, and l,l,3,3-tetra- methylguanidine. The measured thallium-205 chemical shifts at different Ligand/Tl+ mole ratios are given in Table 11. The mole ratio plots for the complexation in various solvents are illustrated in Figure 16. No significant variation in the chemical shift of thallium-205 is observed upon addition of ligand to the thallous salt in the case of tetramethyguanidine which 66 Table 9. Log of Formation constants and the Limiting Chemical Shifts of (DT18C6.Na)+ Complex in Various Solvents at 32 : 1°C. Solvent Log Kf Slim (ppm) Nitromethane 3.25 i 0.15 -4.39 i 0.10 Acetone 2.42 :_0.48 -7.91 :_0.08 Propylene Carbonate 1.87 i 0.10 -7.43 i 0.14 Acetonitrile 1.82 i 0.10 -6.29 :_0.05 Dimethylformamide 1.52 i 0.21 -5.06 i 0.09 Dimethylsulfoxide ~ 0 - Table 10. Log of Stability Constants and the Limiting Chemical Shifts of(DT18C6.Cs)+ Complex in Various Solvents at 32 i 1°C. Solvent Log Kf 611m (ppm) Nitromethane 1.16 i 0.07 16.9 i 9.0 Acetonitri1e(a) 0.97 i 0.27 94.1 i.3-° Propylene Carbonate 0.96 i 0.24 -24.2 :_6.6 Acetone 0.61 i 0.09 65.2 i 3.8 Dimethylformamide(a) 0.56 i 0.29 2.10 i 1.27 Dimethylsulfoxide 0 - (a)Reference (119). 67 Seao.o n voaUHB mo coflumuucmocouc Smo.o n voaUHB mo coflumnucwocouo amoo.o u weaves mo coaumuusmocoon zHo.o u voaoaa mo :oHumuucmocoum Nm.0mm 0H.0 Nm.HNm mv.m mm.MHH ¢N.v 0N.MBN 00.m mm.NvH 0H.¢ NH.mmm Hw.v HN.hHm mm.v 00.NHH hm.m mm.NhN mm.m 00.0vH 0N.m H0.mmm wH.m Nb.00m mm.N mm.HHH v0.m ¢M.th m¢.N mm.FMH v0.N 0v.mmm mm.N 00.00m mm.N m0.HHH Hv.N mo.NhN mv.N 0N.0MH mm.H 0m.vmm Hm.H mm.mNN 00.N mH.0HH mh.H vm.NhN mm.H bh.mMH MN.H NB.VMN 00.H HN.HON mm.H vm.0HH mm.0 N0.VNN mm.H N0.MMH 00.0 0H.hmm 0N.H 0H.0HN 0H.H 5h.0HH 00.0 m0.mmN 0N.H HN.MMH 05.0 vm.0mm 0H.H 0m.hhH 00.H mv.00H hm.0 Nm.NNm 00.H mH.NMH om.0 hm.hmm m0.H m0.mHH eh.0 m0.00H 00.0 Hm.vHNl 00.0 0m.0MH 00.0 Hh.vmml 00.0 0H.00Nl 00.0 Assess +aa\e Assess +He\e Assess +Hs\q iseeve +He\e Assess +Ha\q @029 02¢ thQ Q22 MUd muco>aom msomsvmcoz moowum> CH momaan mo mocwmmnm wnu Ga VOHUHB MOM mama amenm HMUflEmso I oflumm mac: .HH manna 68 “370 “250‘ ‘130 A -l0 8 3- llo H” — : ; ; TMG 00 DMF 230 ._._ 4 NM 0 :A. .4; AH“ 350 —- " “C I I I I I I I 0.00 I.0 2.0 3.0 4.0 5.0 6.0 7.0 [DTI8C6]/[Tl*] Mole Ratio Thallium-205 chemical shifts vs. DTlace/Tl+ Figure 16. mole ratio in various solvents. 69 would be expected to have a high donor ability. On the other hand, in solvent of weak donor ability such as nitromethane (DN = 2.7), the resonance frequency of the 205T1 is strongly affected by addition of the ligand, and the mole ratio plot consists of two straight lines inter- secting at 1:1 mole ratio. This behavior indicates the existence of a very strong interaction between the thallium ion and the ligand. In such cases the formation constants cannot be determined and we can only conclude that the log Kf > 5. A gradual shift is observed in dimethy- formamide solutions which could be due to the formation of a weak complex. It is interesting to note that in acetonitrile solutions as the DT18C6/Tl+ mole ratio increases from O to 1 the 205T1 resonance shifts upfield, then further addi- tion of the ligand gives a downfield shift. Beyond the mole ratio of 2, the resonance seems to level off to a constant value. This behavior clearly indicates a two- step complexation reaction, first the formation of a 1:1 complex and then the addition of a second molecule of the ligand gives a 2:1 (ligand:metal ion) sandwich complex. The formation constants for the (DT18C6-T1)+ complex in various solvents are listed in Table 13. The results show that the variations of the solvent can pro- duce a significant change in the binding properties of the ligand. 70 4. Complexation of the Tl+ ion with 18C6 Table 12 summarizes the results obtained in the study of complexation of the T1+ ion by lB-Crown-G in various non-aqueous solvents. The plots of the thallous ion chemical shifts versus the 18C6/Tl+ mole ratio are shown in Figure 17. Depending on the nature of the solvent, the thallium-205 resonance shifts upfield or downfield with the increasing concentration of macrocyclic ligand. The more pronounced break in the DMF curve, compared to that in the DMSO curve, is a good indication of a stronger polyether complex formation in DMF than DMSO solutions. The behavior in acetone solutions shows that two com- plexes may be formed. Indeed only the slight upward drift of the chemical shifts beyond the 1:1 mole ratio can be used as an evidence for the formation of 2:1 complex, but the data do not permit determination of K The data 2. collected in Table 13 show that the stability constants of the complex are much lower in dimethylsulfoxide and dimethylformamide solutions than in nitromethane, acetonitrile and acetone solutions. These results re- flects the much stronger cation solvation by the DMSO and DMF molecules, compared to the other solvents. 5. Complexing of the Li+ ion by TT12C4 The results of our lithium-7 NMR studies for the interaction between Li+ ionenulmacrocyclic ligand, 1,4,7- trithia-lZ-Crown-4, are tabulated in Table 14. The lithium-7 chemical shifts as a function of TT12C4/Li+ 71 v .Smoo.o u voaUHB mo scasmsscmosouo izee.e n scaoee mo soaumsucmocooe izme.o u caves mo coaumsusmocoom ma.mcn mm.e ao.am- as.c m~.mmI se.m ma.~eu em.a mm.aaeu mm.s ma.mmu Hm.e e~.meau mc.c co.a~- as.m He.eeau me.a ~m.mHI se.m ce.sau eo.a mm.maau am.m sc.eI ae.~ He.aeHI ee.e mm.mau ea.m aa.~man cm.~ am.a mc.~ aH.eeHI ec.m ea.aean ma.m Ha.aau ae.e c~.~maa mm.~ se.m~ ca.~ m~.ecau mc.m ma.cean am.~ me.eau ea.m sc.meHI ec.~ am.mm em.a Na.ecau ea.~ cc.ceeu ae.e ae.aau ea.~ ea.aean ea.e ee.me ac.a am.acHI mm.~ ac.csan ~c.a ac.meu ae.a ac.aaan ~e.a ac.ma cm.a Hm.mcau ea.a as.ceau mH.H ea.msu ma.e as.amHI mc.a Ha.mma ao.a ea.eees ee.e ac.aea- me.a em.mea ee.a mm.eaeu ma.e ea.mea ea.e ee.eee- e~.H e~.mman Ha.e am.eeu as.e ma.meaI as.e ec.mca cm.e ec.eaHI ~e.a ae.cmaI me.e am.e ea.e He.eman mm.e cm.~em me.e ea.esau ma.c ea.meau ec.e a~.as em.e em.maau ee.e ea.am~ cm.e He.aaau mm.e c~.eaHI em.e me.eea sa.e ae.aaau aa.e aa.am~ m~.e as.eem- ec.e Hm.eamu ee.e me.ame ee.e se.ee~I ee.e ee.e~m ee.e fiancee +He\e Assess +Hexs fleeces +Hexe Achebe +Hs\e Leases +Hexe o z¢ nmzn a momzo muam>aom m50aum> cw cow +HB nufls memEOU coma “Ow mumo umwzm Hmowfiwnu oflumm 0H0: .NH magma 72 'fl50 . : c 1 —: . DMF '50 r— I - : ““ DMSO 50 l ISO. 250- 350-— ‘— “370 8(ppn0 -330 -290 -250 -2ufi _ _ . t—=AC '470 NNI I 0 : AN 1 I I I I I I Ii. 0.00 I0 20 3.0 4.0 5.0 6.0 7.0 8.0 [l8C6]/ [TV] Mole Ratio 'fl30- Figure 17. Thallium-205 chemical shifts vs. 18C6/Tl+ mole ratio in various solvents. 73 Table 13. Stability Constants of Thallium (I) Complexes with Dithia-18C6 and 18C6 in Various Non- aqueous Solvents Log Kf Solvent DT18C6 18C6 Nitromethane > 5 > 5 Acetonitrile > 5 > 5 Acetone 2.97 i 0.01 > 5 Dimethylformamide 1.24 i 0.01 3.35 i 0.06 Dimethysulfoxide ~ 0 1.92 i 0.01 Tetramethylguanidine ~ 0 - 74 oo.e oa.m cm.~ oo.m ee.HI aa.~ He.e om.m em.m om.~ om.o cc.~ so.HI cc.~ cm.e om.m He.~ m~.~ om.o om.m ma.ou cm.~ em.o oe.~ mm.~ oo.~ a~.o eo.~ em.eu mm.a ma.e om.H e~.~ se.a mm.o om.H em.eI m~.H oe.a mN.H em.~ o~.H m~.e e~.H cm.eI HH.H mm.o oo.H ce.~ mm.o m~.e eo.H cae.- mm.o ~o.a mm.e He.~ mm.o -.o ma.o cm.oI ma.o oo.H oa.o cm.~ om.e -.e am.o cm.on cm.o mm.o om.e me.m mm.o -.o mm.o mo.HI mH.o se.o mm.o ma.~ ov.o ca.o eo.c cm.on oe.o mm.o oo.o oe.~ eo.e “suave +a0\e Aeeeve +aqxe Random +aq\e Aeeewm +a0\q 22 Oman 04 we v .Uoom um mucm>aom msoflum> cw comaaa mo monommnm 0:» Ce Az~o.ov oaUHA How mumo unflnm Hmofl50£UIofiumm 0H0: .vH manma 75 ac.c- ce.m mc.cn em.m mm.~- om.~ oc.e- oe.m am.m- ee.~ No.0- em.e am.~- em.e em.e- em.~ mc.eu mm.e mm.~- ma.a ee.e- oe.m ac.eu ee.a as.~- eo.a mm.eu cm.a am.c- ms.e as.~- mm.e ss.e- oe.e He.cu em.e ae.~I ms.o Hs.eu em.e ec.e- em.o ea.~- em.c Na.e- eo.e ec.oa ec.e am.~- ee.c Assess +ae\e Leases +aexq Aeeeve aexe has um 24 Ammscaucoov as cache 76 mole ratios in seven nonaqueous solvents are plotted in Figure 18. It is seen that in all cases, the frequency of the lithium-7 resonance is almost independent of the Ligand/Li+ mole ratio which suggests that the immediate environment of the lithium ion is not changed upon addi- tion of the ligand. It is evident, therefore, that the interaction between lithium ion and TT12C4 is not appreciable. At least two factors play an important role in causing this behaviour. The first is the large size of the sulfur atom compared to the oxygen atom, which results in a decrease in the ligand cavity size and in an alteration of the ring conformational energies. On the other hand, complexing of Li+ is weakened appreciably by the sulfur substitution in the polyether ring due to the smaller negative charge on the donor atom. In addition, the Li+ ion is relatively largely solvated and it may be more difficult to strip the solvent molecules from the Li+ ion than from the larger cations. 6. Complexation of the Ag+ ion by TT12C4 Complexing of silver ion by TT12C4 was studied in acetonitrile-d dimethysulfoxide-d and pyridine solutions 3' 6 using proton NMR technique. The chemical shifts for three kinds of hydrogen a, b, and c (Figure 19) of the thia- crown ether were determined relative to the TMS as an internal standard. The results of these studies are given in Table 15 and the plots of the 1H chemical shift as a 77 .muc0>H0m .mH messes msomsqmcoc :H oflumn maoe +flq\¢UNHBB .m> muwfinm HmoflEmno nIEDflsuHA r3? Tom_..._:_...\..¢._ .3 as E .... a... 1 . _ J I z< [I .19.”... .vN- .00 m, OMS—O ...“... Ill .. - s- w an: ( o II 3:401 I c.— ?~ ?a 78 Table 15. Hydrogen Chemical Shift Data for TT12C4 in the Presence of Ag+ ion in Various Solvents Solvent Ag+/L 5Ha (ppm) 6Hb (ppm) 6H0 (ppm) AN-d3(a) 0.00 2.77 2.63 3.63 0.50 2.88 2.75 3.58 0.78 2.90 2.77 3.55 1.26 2.93 2.80 3.52 1.52 2.92 2.80 3.52 1.94 2.92 2.80 3.50 2.50 2.93 2.80 3.52 3.40 2.92 2.80 3.50 Py(b) 0.00 2.83 2.60 3.50 0.48 2.87 2.67 3.52 1.10 2.90 2.70 3.53 1.60 2.90 2.70 3.53 2.00 2.90 2.72 3.53 DMSO-déa) 0.00 2.77 2.60 3.55 0.50 2.85 2.72 3.50 0.87 2.90 2.78 3.48 1.00 2.90 2.80 3.48 1.14 2.92 2.82 3.48 1.52 2.92 2.82 3.48 2.18 2.93 2.83 3.48 2.50 2.93 2.83 3.48 3.00 2.93 2.83 3.48 (a) (b) Concentration of TT12C4 Concentration of TT12C4 0.05M. 0.10M. 79 b/‘f { S 0 2(5 S]; U a C1 Figure 19. Structural formula of 1,4,7-trithia-12-Crown—4 of Ag+/TT12C4 mole ratio for various kinds of hydrogen are shown in Figure 20. The chemical shifts for different types ofihydrogen (a, b, and c) are falling in the order of Hb > Ha > Hc' and the most pronounced shifts are ob- served for the Hb protons. These data clearly show that there is a definite interaction between the Ag+ ion and TT12C4. Chemical shifts observed for the various protons may be the result of small differences in their magnetic environment which in turn are sensitive to conformational changes. C. Comparison of the Results The results obtained in this work, demonstrate the weaker complexing strength of l,lO-dithia-lB-Crown-é than 18-Crown-6 for the sodium, cesium and thallium ions (Table 16). Two factors seem to be important in causing this loss of stability. The first is the larger size of the sulfur atom compared to the oxygen atom, which results 80 3.60 3.50 - Hc 3.40 — \\ 4‘\\ 2.90 — 8 (ppm) 2.80 2 .70 2.60 00 I0 2.0 3.0 [09+] /[TT l2C4] Mole Ratio The variation of the proton chemical shift of the three different protons of TT12C4 with the Ag+/TT12C4 mole ratio in DMSO-d6. Figure 20. 81 .mnmn cm>flm mum HM mo monamp on» maco “mwfiuow mum mmmemEoo quo mam Amu nuom .mm mocmnmmwmfinv . GOGGHG 0 He.cH~m.H ee.euee.m ee.cuae.a 02 oz 02 mcaxOmHsmHssumEae ce.eHnm.m ms.oumm.m ee.eum~.~ He.euam.e a~.eucm.o H~.oumm.e mcaEcEsomamsumEHe m A m A em.euem.m m A s~.ousm.o ea.eH~e.H meannesoumom I ma.euae.e a A I e~.eucm.e Ha.eusm.a mumcoesmo msmaaeosm m A m A a A Nc.eHma.~ me.eHHe.e ma.eume.m 0coumo¢ m A I a A m A so.eHeH.H ma.eHm~.m msmcumeosuaz +HB Anv+mo Amv+mz II +HB +mo . +mz mmmwmmm some mosses mm ace musm>aom m50anm> am mnmnum :3OHU mowa 0cm womaaa suwz mwxwamfiou AHV Edwaamne mam .Enwmmo .Eflwcom 03¢ mo musmumcou huwawnmum .mH magma 82 in a decrease in ligand cavity size. The second is lower electronegativity of the sulfur atoms compared to the oxygen atoms, which results in a decrease in electrostatic interaction between metal cation and the donor atoms. Complexation studies of the cesium ion with 18C6 were performed by Mei (36) in various nonaqueous solvents. Both 1:1 and 2:1 (ligandzmetal) complexes are formed in all solvents studied. The weakest complex was found in DMSO solutions, K = 1100 i.100 and K = 1.00 i 0.4. Complexing 1 2 of Na+ by 18-Crown-6 was studied by Lin (120) in various organic solvents. For comparison purposes, the results of this work and Mei's studies are included in Table 16. The comparison of our results obtained from sodium- 23 and cesium-133 NMR studies on the complexation of Na+ and Cs+ ions by DT18C6 in various solvents with Mei's and Lin's results show some evidence of stronger interactions of these ions with 18C6 than with DT18C6. Although DT18C6 prefers binding with the thallium (I) ion rather than the alkali metal ions a decrease in the complexation strength is still observed when the sulfur substituted ligand is used. Pearson's hard-soft acid-base theory (121) does not seem to explain the complexation strength observed for the thallium ion. According to the Pearson's theory, the soft acid, thallium ion, should prefer to bind to ligands containing sulfur atoms (soft base) rather than ligands possessing oxygen atoms (hard base). From size 83 consideration, however, the thallium ion may prefer to form a stronger complex with 18C6 than with DT18C6. It is probable, therefore, that a primary reason for the lower stability of the complexes formed between DT18C6 and thallium ion, is that this cation is too large to fit into the ligand cavity since the hole in the macro- cycle is smaller when the sulfur atom is a donor than when the oxygen atom is a donor (10). It would be expected, therefore, that the metal ion does not fit into the cavity but it may lie above the plane of the ligand ring. On the other hand, the cavity of 18C6 does accommodate the Tl+ ion because its ionic diameter (2.80 A°) is close to that of the 18C6 cavity. Comparison of the data obtained from thallium- 205, sodium-23 and cesium-133 NMR studies in various sol- vents (Table 16) shows that the stability of the appropriate complexes is falling in the order (DT18C6°T1)+ > (DT18C6-Na)+ > (DT18C6-Cs)+ regardless of the solvent in which the complexation takes place. This seems reason- able because the thallium ion is capable of forming ionic bonds with the oxygen atoms and (partially) covalent bonds with the sulfur atoms, whereas the sodium and cesium ions form only ionic bond. An exception, however is observed in the case of complexation of Na+ and Tl+ ions by DT18C6 in dimethylformamide solutions in which the stability of the (DT18C6-Tl)+ complex is lower than that of the 84 (DT18C6-Na)+ complex. A possible reason for such an ex- ception may be due to the poorer solvating ability of dimethylformamide for the sodium ion than for the thallium ion. Since dimethylformamide posseses a nitrogen atom (soft base), it may prefer to solvate the Tl+ ion (soft acid) rather than the Na+ ion (hard acid). Although the sodium ion (higher charge density compared to the Cs+ ion) is probably more solvated than the cesium ion, it forms a stronger complex with DT18C6. Probably the lower stability of the (DT18C6-Cs)+ complex is due to the fact that the cation is too large (cation diameter = 3.34 A°) to match the ligand cavity. On the other hand, the cavity size of the ligand may accommodate well the sodium ion (cation diameter = 1.90). The results obtained with lithium-7 NMR studies on the complexation of the Li+ ion by TT12C4 in acetonitrile, dimethylsufloxide, tetrahydrofuran, propylene carbonate, nitromethane, acetone, and pyridine solutions show that the interaction between lithium ion and TT12C4 in these solvents is not strong. Complexation of lithium ion with 12C4 in various solvents was studiedkurSmetana (117) using lithium-7 NMR technique. Some of his results are given in the following Table. As we expect, comparison of these data with our results shows that the interaction between Li+ with 12C4 is much higher than the interaction of this ion with TT12C4 in various nonaqueous solvents. 85 Table 17. Stability Constants of (Li-12C4)+ Complex in Different Solvents Solvent log Kl log K2 NM 4 l . 57 AN 4.25 AC 1.62 PY 0.70 DMSO 0 These results are not unexpected, since the cavity size of TT12C4 is so small that a metal ion could not be located in it. Likewise the electronegativity of the oxygen is higher than that of sulfur which makes the C—S bond less ionic than the C-0 bond. Comparison of the data from the proton NMR studies on the complexation of the silver iontanT12C4 is AN-d3, DMSO-d6 and pyridine with the results of lithium—7 NMR measurements shows some evidence of stronger interaction between silver ion and TT12C4 than the interaction of lithium ion with this ligand in nonaqueous solvents. This seems reasonable because the Ag+ ion is capable to form ionic bond with oxygen atom and more covalent bonds with sulfur atoms than the lithium ion. Therefore the covalent bonding plays an important role in the Ag+ complexes of polyethers containing sulfur atoms. In general substitu— tion of sulfur atom for oxygen atom in the polyether ring reduces affinity for Tl+ and alkali metal ions but greatly increases the binding to silver ion. CHAPTER IV THERMODYNAMIC STUDIES OF THE COMPLEXATION OF DIBENZO-27-CROWN-9 AND DIBENZO-24- CROWN-8 WITH CESIUM ION IN VARIOUS NONAQUEOUS SOLVENTS 86 A. INTRODUCTION While the properties of macrocyclic polyethers and their alkali complexes have been under active investigation during the past several years, it seems, however, that most of the attention has been focused on alkali cation com- plexes with small crown ethers and, in particular, on 18— Crown-6 and its derivatives. Few thermodynamic studies have been reported for the reaction of metal ions with large macrocyclic crown ethers (i.e., larger than 18-Crown- 6) despite the very interesting properties of these ligands. Therefore, little is known of their complex forming ability. Izatt and co-workers (9, 122) determined the thermodynamic quantities (AGE, ASS, and 0H3) for the complexes of alkali and alkaline earth cations with 21-Crown-7, dibenzo-24- Crown-8, and dibenzo-27-Crown-9 in methanol or methanol- water mixtures by calorimetric technique. The present study reports the use of the cesium- 133 NMR study of the Cs+ ion complexation by large macro- cyclic crown ethers, dibenzo-27-Crown-9 and dibenzo-24- Crown-8 in various nonaqueous solvents, and shows how the thermodynamic parameters for the complexation process is affected by the nature of the medium, the cavity size, and the number of donor atoms in the polyether ring. 87 88 B. Complexation of Cs+ ion by DB27C9 a. Effect of the solvent on the stability of the (DBZ7C9°Cs)+ complex Cesium-133 chemical shifts were measured as a func- tion of DB27C9/Cs+ mole ratio for the (DB27C9-Cs)+ system in dimethylsulfoxide, dimethylformamide, methanol, pro- pylene carbonate, acetonitrile, pyridine, acetone and nitromethane solutions at 30°C. The cesium-133 NMR chemical shift-mole ratio data are given in Tables 18-26. The 133Cs chemical shifts as a function of variation of the the DB27C9/Cs+ mole ratio in various solvents is shown in Figures 21 and 22. The chemical shift-mole ratio plots were analyzed by a computer fit of the data to the equation relating the observed chemical shift to the formation constant of the complex as described in Chapter III. The stability con- stants and the limiting chemical shifts for the (DB27C9-Cs)+ complex in various solvents at 30°C are listed in Table 27. A change of solvent would be expected to result in changes in the complex stability, especially when the cation and ligand are strongly solvated. As mentioned in the first chapter, the solvation of the ligand and metal cation are influenced by the donor ability and dielectric constant of the solvents as well as by the shape and size of the solvent molecules. However, it has been shown that 89 the donor ability of a solvent plays the most important role in the behavior of alkali complexes in nonaqueous solvents (16). It is seen that (Table 27) the formation constants of the (DB27C9-Cs)+ complex are much lower in dimethyl- sulfoxide and dimethylformamide solutions than in the other solvents. Dimethylsulfoxide (D-N = 29.8) and dimethylformamide (DN = 26.6) with high donor abilities can solvate Cs+ ion quite strongly and, therefore, compete with the ligand for the cesium ion. Therefore in these solvents the formation of the (DB27C9-Cs)+ complex is weakened. These results are consistent with the previous studies of the solvation of the Cs+ ion by different sol- vents which showed that DMSO and DMF are much more strongly solvating solvents than the other solvents used in this investigation (11). It is interesting to note that although DMF and NM have comparable dielectric con- stants (Table 27), the formation constant of the complex in NM, a poor donor solvent, is much higher than DMF which has a larger donor number. Pyridine seems to be an exceptional solvent. Al- though it has a high donor number (DN = 33.1), the forma- tion constant of (DB27C9-Cs)+ complex in this solvent is surprisingly high. Similar behaviors for the pyridine solutions of alkali cation-crown ether complexes have been observed by Mei (11), Shamsipur (70), and Smetana (117) 90 in our laboratory. A possible explanation for this excep- tion may be the relatively poor solvating ability of pyridine for alkali cations. Since pyridine is a nitrogen donor and, therefore, a "soft base", it may not solvate strongly the alkali metal ions which are relatively "hard acids" (121). b. Thermodynamic parameters for the complexation of Cs+ ion by DB27C9 In order to have a better understanding of the thermodynamic behavior of (DB27C9-Cs)+ system, we con- ducted the complexation studies of Cs+ ion by the macro- cyclic ligand, DB27C9, in various nonaqueous solvents at different temperatures. The chemical shifts of the cesium-133 resonance were measured as a function of DB27C9/Cs+ mole ratios at various temperatures in dimethylformamide, propylene carbonate, nitromethane, acetonitrile, acetone, pyridine, and methanol solutions. In all cases studied here, only one resonance of the metal ion was observed (regardless of the Lig/Cs+ mole ratio and the temperature), which indicates a fast exchange be- tween the two cationic sites (i.e. free ion in the bulk solution and the complexed ion). The data obtained from these studies are given in Tables 18-25. Figures 23—30 show the variations of the cesium-133 chemical shifts as a function of DB27C9/Cs+ mole ratios at different temperatures in the above 91 Hw.va va.~vI mo.HvI vQ.OVI mo.ch mm.mmI ov.m mo.va mm.va mN.HvI Hm.ovI mH.ovI mh.mmI mm.m h¢.¢¢| m~.va m¢.HvI hw.ovl ma.o¢I Hw.mmI vo.a mo.m¢I mN.MVI m>.HvI vh.ovI v¢.OVI Hm.mmI wo.a mm.va ww.va om.va wa.avI mm.ovI mo.o¢I mm.H no.th mm.v¢1 Hm.va oo.HvI om.ovI mN.OVI om.H mm.m¢I mm.va mm.m¢I mm.N¢I mm.HvI mh.ovI vo.H H~.omI Nv.th vm.m¢I >>.mmI vm.mv| «H.Nvl mm.o mm.¢mI oo.NmI mm.omI mm.mmI mo.va ma.th om.o ow.mmI Ha.hmI na.mmI hm.mmI mm.NmI hm.HmI wm.o v~.me mm.omI Hm.mmI mm.hmI nm.mm| H>.mmI mH.o om.bml hm.mmI mm.moI mh.HmI mm.omI ma.mmI oo.o om me am om ov om Us wusumummfima +m0\q .mwusumummfime msoflum> um 22 CH mohmma mo mosmwmsm 0:» ca Lamoo.ov zommo ecu memo umegw Hmoasmno I caumm 0H0: magma 92 0H.mHI mH.mHI om.mHI Hm.oml om.HNI no.HNI mm.m om.mHI om.mHI m©.hHI oa.mHI wm.mHI 5N.ONI mm.m mm.mal mm.vHI mm.mHI 00.5HI ow.mHI cm.mHI mm.m mm.NHI vw.MHI 05.0HI mm.wHI mh.hHI wm.mHI «H.N 0N.HHI hm.HHI mm.NHI Hw.vHI mm.mHI mm.mHI mm.H em.m I mv.OHI mm.0HI MH.NHI mm.mal mm.vHI vN.H mo.m I oa.m I av.m I mm.oal hm.HHI mN.NHI oo.H ov.h I Hm.h I mh.h I wm.m I mm.m I v0.0 I vm.o Ho.h I Ho.h I mo.h I mm.> I Ho.m I mo.m I Hh.o mm.m I mm.m I no.m I mm.¢ I mm.v I Hm.¢ I mv.o om.v I mh.m I ma.m I wo.m I ov.H I v0.0 I hm.o mv.m I mH.H I mN.o NN.N mm.v vm.m 00.0 ow om om OH o CHI +mU\A Uo musumquEoB .mmnsumu0e50e usmumwmao as use as mommmo mo mocmmoum may CH AEmoo.oV zumwu How mumn amanm Hmowfimno I owumm 0H0: .mH waame 93 mm.mmI mm.va mw.mmI Hm.MMI mm.mmI «m.NMI NH.m m>.mmI vv.va Hm.MMI mm.mmI mm.mmI mv.~mI mv.m H~.mmI mm.va mm.mmI em.MMI m¢.mmI m~.mmI mm.H mm.mmI mH.mMI ma.¢mI mo.va mm.mmI H¢.NMI hm.H ww.mMI vv.mmI vv.VMI mm.¢mI om.mMI mq.mmI mm.H Ho.hMI ma.omI mo.mMI mm.vMI mm.va mm.NMI ma.o Hm.mmI mo.hMI mm.mmI om.mMI mm.va NH.MMI om.o Ho.mmI mm.hmI mm.wmI mm.mMI ma.mmI mv.mmI mm.o hv.mmI mo.mmI mm.mmI «H.0mI mm.mmI Hw.mMI mv.o nm.oVI ov.mmI Ho.mmI vm.nmI Hm.mmI mm.va wH.o mv.HVI mN.OVI Ho.mmI mm.mMI em.nMI Hm.mmI oo.o mm mm we ov om NH Uo musumu0m809 +mU\A .mwusumnwmfima msoflum> um on CA mubmmn mocmmmse men as figmeo.ov zommo 000 meme umacm Hmoa50co I caumm 0H0: wanna mo.wHI Hm.mHI mo.mHI No.mHI oo.mHI ow.vHI hm.m 94 mh.vHI om.mHI No.mHI om.mHI Hm.val mm.vHI mm.m vm.MHI hm.vHI oo.mHI mH.mHI oo.vHI NN.VHI No.m Hm.HHI mo.MHI mm.mal NN.¢HI va.val Hm.mal mw.a om.oal hm.NHI hm.mal mb.maI mm.mHI vv.mal H0.H mo.m I vw.m I mm.HHI mm.mHI Hm.NHI mm.mHI ON.H mH.m I mm.h I mm.m I H¢.oal mo.HHI em.HHI mo.H mm.m mm.m mo.v mm.m Hw.m mm.m mm.o mm.m Hh.m hm.m mv.m mv.> NH.m mm.o mm.m mm.m mm.oa Hm.oa mv.ma mm.ma mm.o NH.mH mm.ma mn.mm Nh.vm mm.mm mm.mm 0H.o on.mm av.mm om.mm No.0m Ho.vm mm.mm 00.0 mm mm mm om 5H m +m0\a Us musumuma80e .mousumuwaE0e msoflum> um z< ca mommaa mo mocmmwua may Ca AZmoo.ov zommu Mom muma amasm Hmofi50no I oflumm 0H0: .HN maame 95 HH.HMI om.omI mm.OMI mH.OMI av.mNI mm.m Ho.HmI om.ch om.CMI ma.omI H0.mNI vo.m om.omI mh.omI em.OMI mH.omI av.m~I mm.m m¢.omI mH.omI HH.omI mo.OMI mm.mmI om.a nm.m~I om.mNI Hw.mNI mh.mNI vm.mmI mN.H H¢.mNI mH.mNI mo.mNI oa.mmI mo.mNI Ho.H vm.mNI mm.mNI mm.mmI mm.mmo Ho.m~I mm.o mm.mNI om.hmI mm.hmI mm.nmI sh.wNI mh.o mo.hNI om.wNI mm.mNI mm.mmI vv.v~I nm.o mo.mNI mm.vm| mm.¢mI hm.mmI m¢.N~I hm.o om.v~I Hm.mNI m>.NNI hm.HNI av.oml om.o om.mmI NH.NNI om.HNI mn.mHI mo.mHI oo.o mm mm ov om om Us muaumuwaEmB +mo\a mensumhme50e ucwnmamaa um om ca mommmo mo mocmmmua 030 CA AZmoo.oV zummu MOM muma UMfism HmOHEmnu I ofiumm 0H0: magma 96 hm.mHI nm.mHI «5.0NI mm.oml mo.oml om.mHI mm.m mb.hal em.mHI mm.oml mm.mHI om.mHI vm.mHI mh.N HB.©HI m©.mal mm.mHI mm.mHI hm.mHI mH.mHI mN.N mm.val HB.©HI mH.mHI vw.wHI om.mHI hm.mHI vw.a Nv.OHI v¢.mal mH.mHI No.0HI Ho.hHI No.wHI om.a mo.m I hm.HHI mm.MHI hh.vHI on.mHI mo.hHI mo.H mm.m I mh.m I mm.oHI mm.HHI mm.NHI NN.VHI mm.o no.v I no.0 I No.m mo.m I wv.m I mm.oal 05.0 av.m «v.0 vm.n mm.w wh.m NN.HH om.o mN.mH mm.hH mh.om H©.NN mo.vm om.mN MH.o Hm.om mH.NN mh.vm mm.hm m¢.mm vm.mm oo.o mm mm m0 mm mv om +m0\q Us wusumumasoe .mwusumumaEmB msowum> um ya me mmeaa mo wocwmmua may Ca AEmoo.ov ZUmmU Mom muma umwnm HmowfiwnU I oflamm 0H0: .mm maama .2moo.o H ZUmmU MO COHDmHuaanOU "UUOZ 97 mH.oql mm.wmI mo.mml mh.hml mm.hml oa.m mm.ovl mo.mMI HN.wmI mN.NMI 0N.hml mm.m oo.HVI mm.mm| mm.mMI Hm.hml mv.wml mn.a mo.NvI Uv.owl mv.mml 0v.mml v5.5ml hN.H HB.NvI Ho.HvI mm.mMI mo.mml Hm.mml mo.H om.mvl Ho.mvl mm.ovl mm.mmI mw.mml mm.o mm.vvl No.va HB.H¢I 55.00I mh.mml mh.o mm.mvl mm.vvl vo.mvl mm.HvI 0H.Hvl mm.o mm.NVI wh.mvl NH.va HH.mVI mH.va hm.o mo.va mm.NVI mo.wVI mv.mvl mm.vvl NN.o hm.HmI mv.oml mm.mwl mm.hvl hw.mvl 00.0 mm om ow om mm +mU\a Us waaumumaEmB .mmusumumafima msoflum> um Hocmnumz CH mUtha xa +mU mo coHumxmaaEOU.u0m muma umanm HmoflEmaU I Owumm 0H0: .vm magma 98 .ocmmHH mo coflusaow pwumuaumm any .mac.o I zommo mo .ocooisv va.mmI ow.mmI hm.mmI mo.wmI Aavam.m va.mMI oo.>mI mm.UMI mo.mmI mm.~ hm.mMI ow.>mI ma.hmI mo.mmI mm.H mm.mmI mo.mmI mm.>mI om.mmI mv.a mm.mMI Hm.mmI mo.mmI mm.mmI no.a mo.o¢I om.mmI nm.mmI mo.>mI mm.o vm.ovI mm.mmI vw.mmI mm.hmI nm.o 5v.HMI on.ovl Hm.mMI mm.mml mh.o mm.va mo.va mh.HvI mN.ovI mv.o Hm.>vI ma.me Hm.va om.va ma.o ov.va mm.va mm.o¢I om.va oo.o he ow Hm ma Us wuaumuwafime +mU\a m0 mmusumuoasma m50aum> um Hocscumz ca moammo an coH so so coausxsa 200 Mom sumo mmHIesammo + .mN manmfi 99 Table 26. Mole Ratio Study of DB27C9 with CsSCN (0.005M) in DMSO at 30°C L/Cs+ 6(ppm) 0.00 65.04 0.21 63.18 0.46 61.16 0.75 59.23 1.04 57.13 1.17 56.18 1.46 54.10 1.81 51.86 2.29 49.23 3.02 45.19 100 Table 27. Formation Constants of the (DB27C9-Cs)+ Complex in Various Solvents at 30°C. Solvent D-N(a) D(b) Log Kf Dimethylsulfoxide 29.8 45.0 1.38 i 0.03 Dimethylformamide 26.6 36.1 2.20 i 0.01 Propylene Carbonate 15.1 65.0 3.64 i 0.03 Acetonitrile 14.1 37.5 3.89 i 0.03 Pyridine 33.1 12.3 4.15 i 0.02 Acetone 17.0 20.7 4.24 i 0.15 Nitromethane 2.7 35.9 4.29 i 0.04 (a) (b) D’N = Gutmann Donor Number. D = Dielectric Constant. 101 60 50 DMSO 3N " zu - E a. O. V 60 In - ° -10» K ‘ .:‘-*' 1...,f- AJU - DMF -20,— 4; FY -‘~‘ I J l l 1 0-0 In 2-0 3-0 4-0 5-0 [0 B 2709]/ [05+] Figure 21. Cesium-133 chemical shift vs. (DB27C9)/(Cs)+ mole ratio in various solvents. 102 -18 —24 .— ‘3" ' M #0— AC P1: MeOH if NM L J I J 0-0 In ,_ 2-0 3.0 4.0 5.0 [D B 27 09]/[Cs+] Figure 22. Cesium-133 chemical shifts vs. (DB27C9)/(Cs+) mole ratio in various solvents. 103 .wmnsumuwa80u ucwnommam um aza Ca cause sacs i+movxxaoa~mos mo coauocsu s as coach assesses mmaIesamso .mm cusses To“; \ T383 00.0 00:? 00m 00. N 00.. 00.0 a _ _ a _ a- . /. awn" IlllIIIJIIIIIIIIIIIII .Il mwTI mw. n3? _ . I. c... ...... l.//. 0.. /,.. d . I Q! ..nwu ///. ( /. NI 104 - ‘I2° E" §§° 57 A 64 S 3 -37 . . ' 60 . -38 {59’ -40 “4| 7 I I I 0.0 I.0 2.0 3.0 [082709]/[Cs+] Figure 24. Cesium-133 chemical shift vs. (DBZ7C9)/(Cs+) mole ratio in PC at different temperatures. 3 Of l33C$ (ppm) —> ' III . 45 \ I I I I 0.00 I.00 2.00 3.00 4.00 [DBZTCQ] / [05+] Cesium-133 chemical shift vs. (DBZ7C9)/(Cs+) mole ratio in NM at various temperatures. Figure 25. 0 .00 I .00 2.00 3.00 47.00 [082709] / [0s+] Figure 26. Cesium-133 chemical shift vs. (DB27C9)/(Cs+) mole ratio in AN at various temperatures. -l8 \ 107 ~20\ -22I v~-24\ E Q 3 I0-26 '- . \ '2 H— 0 -28 - \ m \ £8 -30 —- \ 1.. 30 4O —=0- 50 * 55 -34 _.... I I I I 0.00 LOO 2.00 3.00 4.00 [D 82709]/[0s*] Figure 27. Cesium-133 chemical shift vs. (DB27C9)/(Cs+) in AC at various temperatures. 108 04 O DBZ7C9 + 0s+—>Py N O / ,// E IO .1“ o- I 3. \ s o ... «P '2 0 "lo ... 40 -20_ __ I I l I I.00 2.00 3.00 4.00 [082709] / [05*] Figure 28. Cesium-133 chemical shift vs. (DBZ7C9)/(Cs+) mole ratio in PY at various temperatures. 109 -37 _ 25 *0 30 40 "ESS>"' 5K) 59 -‘;ES_—' g. ‘45 Q ‘0 , I I I I O .0 LG 2 .O 3.0 4.0 [08 27 09]/[Cs+] Cesium-133 chemical shift vs. (DB27C9)/(Cs+) Figure 29. in MeOH at various temperatures. ’36 — l6 °C 3| 40 '38 — 47 ..42 — S. -44 3 00 -50..— I I I I 0.0 I.0 2.0 3.0 4.0 [082709]/[Cs+] Figure 30. Cesium-133 chemical shift vs. (DB27C9)/(Cs+) mole ratio in MeOH at various temperatures. 111 mentioned solvents. It is evident that as the tempera- ture increases the plots show less curvature which in- dicates the formation of a weaker complex at higher tem- peratures. This behavior shows that in these solvents the reaction between cesium ion and the ligand is exothermic. The cesium—133 resonance shifts upfield or downfield (de- pending on the particular solvent) with increasing the concentration of ligand and a clear break at a 1:1 (metal ion:ligand) mole ratio is observed. The formation constants of (DBZ7C9-Cs)+ complex in different solvents at various temperatures were com- puted by the KINFIT program (page217) and the results are given in Table 28. The data show again that the stability constant increases as the temperature is lowered. The thermodynamic parameters, AGE, AH;, and As; values were calculated from variation of the equilibrium constant with temperature using the following well known expressions: _ "AH° AS° 1n Kf - RT + T (1) AG° = -RT ln Kf Van't Hoff plots of 1n Kf versus % for the (DBZ7C9-Cs)+ system in various solvents are shown in Figure 31. The enthalpies (ABS) and the entropies (Asa) of complexation were obtained from the lepes and inter- cepts of the plots and the results with the AG; values 112 Table 28. Formation Constants of (DB27C9,Cs)+ Complex in Various Solvents at Different Temperatures Solvent Temperature (°C) Log Kf Dimethylformamide (DMF) 40 2.05 i 0.01 ' 30 2.20 i 0.01 20 2.33 i 0.01 10 2.58 i 0.01 0 2.78 i 0.01 —10 2.89 i 0.01 Propylenecarbonate (PC) 64 2.84 i 0.05 57 2.95 i 0.03 46 3.18 t 0.03 40 3.27 i 0.05 30 3.64 i 0.03 12 > 4 Nitromethane (NM) 90 3.46 i 0.01 75 3.48 i 0.05 61 3.81 i 0.02 50 4.14 i 0.03 40 4.24 i 0.06 30 4.29 i 0.04 Acetonitrile (AN) 77 3.09 i 0.02 63 3.30 i 0.02 45 3.63 i 0.01 30 3.89 i 0.03 17 4.17 i 0.04 9 4.24 i 0.04 Acetone (AC) 55 3.22 i 0.04 50 3.62 i 0.14 40 3.88 i 0.19 30 4.24 i 0.15 20 4.43 i 0.15 113 Table 28 (continued) Solvent Temperature (°C) Log Kf Pyridine (PY) 98 3.08 i 0.01 82 3.35 i 0.02 68 3.61 :_0.06 56 3.78 i 0.02 48 3.93 i 0.02 30 4.15 i 0.02 Methanol 59 3.22 i 0.04 (Conc. of CsSCN = 0.005M) 50 3.32 i 0.02 40 3.46 i 0.03 30 3.52 i 0.09 25 3.74 i 0.02 Methanol 47 3.35 i 0.02 (Conc. of CsSCN = 0.011M) 40 3.40 i 0.02 31 3.51 i 0.04 16 4.09 i 0.16 114 ‘0 r O I- 0 i'° ' I: .I 7 s I *NM 0.- OPV OAN DPc fiAC OMeOH 1 I J 1 4 4 1 I 4 21 an 29 so u 32 33 34 35 1...? T Figure 31. A plot of 1n K vs. % for the complexation of Cs+ ion with DB27C9 in various nonaqueous solvents. 115 (at 30°C) are listed in Table 29. These data indicate that in all cases the (DB27C9-Cs)+ complex is enthalpy stablized but entropy destabilized, and that both quantities AH; and ASE, are quite sensitive to the nature of the solvent. Of the various factors which contribute to the overall entropy changes (see Chapter I, page 28), it seems that the negative entrOpy contribution from changes in configurational entropy of the ligand upon complexation may be very pronounced in these cases. The macrocyclic polyether, DBZ7C9 (contains 9 oxygen atoms in a 27 atom ring) is a representative of the large macrocyclic poly— ethers, which would be expected to be more flexible than smaller crown ethers. Therefore, a change in the ligand, from a "loose" structure in the free state to a "rigid" structure in the complex state, contributes to the negative entrOpy changes as a dominant factor. Quite different Asa values are observed when the reaction is studied in different solvents (see Table 29) because the rigidity of the free and complexed ligand is different in various solvents. The AH: values are also solvent dependent since the solvation energies of the Cs+ ion, the ligand, and of the complex will be different in different solvents. 116 me.m H mm.s I Ho.o H ma.sI oo.o H mo.mI so.o H a~.e scmcoseousaz oo.m H om.m~I oe.a H om.4HI H~.o H mm.mI ma.o H em.e scouooe mm.a H oa.a I mm.o H ao.oI ao.o H ma.mI mo.o H mH.e scacauao os.a H mm.a I me.o H ma.aI so.o H om.mI mo.o H am.m seausacoosos ma.a H He.omI sm.o H HN.HHI so.o H mo.mI mo.o H ao.m suscocuso occaseouo am.a H mw.m I om.o H oo.mI Ne.o H mm.eI oo.o H mm.m Hocscosz oa.a H HH.MHI em.o H o~.sI ao.o H mo.MI ao.o H om.~ moHEsuomaacoseHo Ase sawmmasoo Asaozmmmo.sv “Wmmmwemmms. Cosome as mos us0>aom .mqu>Hom maowavaoz maoHHm> CH meaEOU +AmU.oU>Nmav COM mamamEmHma UHEmaonHmCB .mN mHamB 117 C. DB24C8 Complex with the Cs+ ion The effect of the temperature on the cesium-133 resonance for the (DB24C8-Cs)+ system was determined at various mole ratios of DB24C8/Cs+ in dimethylformamide and pr0pylen carbonate solutions. All of the cesium-133 chemical shifts measured at different temperatures are listed in Tables 30 and 31. The chemical shift-mole ratio plots are shown in Figures 32 and 33. The formation constants of the (DB24C8-Cs)+ complex at various temperatures are given in Table 32. Plots of ln k versus % for the data in this table are shown in Figure 34 and the corresponding thermodynamic parameters are listed in Table 33. Thermodynamic data, show that in both solvents (PC and DMF) the complex is enthalpy stabilized but entropy destabilized, and the enthalpy and entropy values vary with the nature of the solvent. Similar behavior has been observed for (DB24C8-Cs)+ complex in nitromethane, acetonitrile, acetone, methanol, and pyridine solutions (70). The entropy destabilization of the com- plex may be explained by the ligand configurational entrOpy. Dibenzo-24-Crown-8 which is a large macrocyclic ligand should be flexible in the free state, but when it is com— plexed by the Cs+ ion it becomes more ordered, and con- sequently it loses a high degree of freedom in the complexation process. 118 mH.mHI mw.mHI Ho.hHI 0H.mHI HH.mHI mm.omI om.m mm.MHI mo.¢HI Hm.mHI hh.wHI cm.hHI vm.mHI mo.m mm.HHI NH.NHI MH.MHI mo.vHI mv.mHI Ho.hHI oo.m om.HHI mm.HHI no.~HI m>.MHI oo.mHI o¢.mHI mm.H mo.OHI mH.0HI om.OHI mm.HHI mm.mHI mv.VHI mv.H H>.m I m>.m I No.m I mm.m I om.0HI Hm.HHI 0H.H em.> I mm.> I mm.h I «m.h I mm.m I H¢.m I vm.o vm.n I on.m I mo.m I vm.m I eh.m I ov.h I mm.o ho.o I Hm.m I vm.v I mm.v I om.v I om.m I mv.o mm.v I H~.v I MH.m I mm.~ I vh.H I mm.o I mN.o m>.m I Hm.~ I vo.H I mm.o m~.~ mH.v. oo.o om ov om om OH 0 Us masumuma80e +mU\a mmuaumuma60a maoHHm> um mza CH mUvmaa ma CoH +mU mo CoHumonaEOU How muma DMHCm HmOHfioaU mmHIEaHmmU magma 119 oH.mmI me.~mI NH.HmI mm.HmI ao.omI m~.omI mm.m mv.mmI mo.~mI om.HmI H6.amI mao.mI em.omI oo.m mm.mmI mm.~mI mo.HmI Hm.HmI mm.omI oe.omI mm.~ om.emI am.mmI em.~mI om.HmI mo.omI em.omI mm.H no.4mI so.emI mm.~mI mo.NMI ea.HmI om.oMI ~6.H o~.mmI mv.4mI om.mmI sm.~mI mm.HmI Hm.oMI oa.a 4H.omI Hm.mmI am.mmI NH.mmI mo.~mI o~.HmI ~m.o om.amI om.mmI «H.4mI om.mmI mm.mmI mo.HmI mm.o no.8mI oo.omI oo.mmI Hm.emI em.mmI Nc.~mI os.o om.mmI Ho.amI mm.omI om.mmI mo.emI co.mmI mH.o mm.omI Ho.mmI mo.amI om.omI mo.smI mm.mmI oo.o on em om o4 om on Us onsumuma80e +mU\a monaumumasma mCoHHm> um um cH movmmo an sea 00 mcax0HCEoo now some uMHcm assassco mmHIesHumo .Hm mange 120 -4 2 *3- \ Q \ 3 ac, "I12 _’ . 50 ° "l‘5"' . ‘QCI [c 30 20 _20_ I0 0 -22 I I I I 0.0 I.0 2.0 3.0 4.0 5.0 [08 2408]/[0s+] Figure 32. Cesium-133 chemical shift vs. (DB24C8)/(Cs+) mole ratio in DMF at various temperatures. 121 ‘30 " 20°C 30 40 50 7O S -36 3 00 'NBNB ’ ..‘1C)... ‘42 l I I I 0 I 2 3 4 [082408] / [05+] Figure 33. Cesium—133 chemical shift vs. (DB24C8)/(Cs+) in PC at various temperatures. 122 Table 32. Stability Constants of (DBZ4CB-Cs)+ Complex in DMF and PC at Various Temperatures. Solvent Temperature (°C) Log Kf Dimethylformamide 50 1.89 i 0.01 40 2.02 i.0°01 30 2.15 i 0.03 20 2.32 i 0.01 10 2.46 i 0.03 0 2.65 i 0.02 Propylene Carbonate 70 2.70 i 0.04 60 2.78 i 0.02 50 3.09 i 0.05 40 3.12 i 0.03 30 3.25 i 0.04 20 3.56 i 0.24 123 8 I- 4. 7 [- ¥ 0 6 .. x- I: J 5 I- (naucaxsf’ IN 7c 4 __ O (IIII27C9.0s)+ m DH? 4 (nnzacnmsf’ IN 010' 3 l J l l l l I I J 29 3 O 31 32 33 34 3 5 36 37 38 1 3 :F'XIO Figure 34. A plot of 1n K vs. % for the complexation reaction of Cs+ ion with DB24C8 and DBZ7C9 in PC and DMF. 124 om.m + am.HHI ea.o + Ha.mI oo.o + Hm.eI eo.o H m~.m soscoeuso scsaaooum Ne.o H a~.oHI ma.o H NH.oI eo.o H mo.~I mo.o H ma.m steamesomascumEHe 1.x 0H02\Hmov Awaoz\aoo so Asaoz\aoo so a .mo one Cosomv we Co.omc x ooe ucs>aom Ua UCm aza CH meaEOU +AmU.mUvaav How mmHuHquCO OHEmCmcoEHmCB .mm mHamB 125 D. Comparison of the Thermodynamic Parameters Thermodynamic parameters obtained for complexing of the Cs+ ion by dibenzo-27-Crown-9 and dibenzo-24-Crown- 8 in various nonaqueous solvents are compared in Table 34. The results are consistent with the change which might be expected from an increase in the cavity size with a consequent increase in the number of donor atoms and in ligand flexibility. The large differences between the AHé and ASé values for these systems suggest that the size of the ligand has a profound effect on the thermodynamic behavior of these complexes. Since in these systems the effects of the metal cation and its charge type on the thermodynamic parameters in the same solvent are the same, therefore the differences in the size of the ligand (consequently the differences in the number of donor atoms and flexibility of the ligand) will be effective in these systems. It is interesting to note that in all solvents except acetonitrile and methanol the entrOpy values of the (DB24C8-Cs)+ complex are less negative than those for the (DB27C9-Cs)+ complex. It seems reasonable to assume that the dibenzo-27-Crown-9 is a more flexible ligand (because of its larger cavity size) and it undergoes a sizeable deformation upon formation of an inclusion complex, which results in a more negative ASE value for the complexation reaction. 126 .on COCCHmuom Amy Haa.aHmo.omI mm.oHom.HHI ma.onH.mIJ oo.nHom.mNI oa.HHom.gaI Hm.onm.mI Ce oo.oHom.H I oH.oHao.mI mo.onm.mI mm.HHoa.a I mm.oHoo.oI so.QHma.mI as sm.oHos.H I oH.on~.oI HH.QHHs.mI ma.~Hmm.s I Ho.onc.HI oo.ono.mI zz mo.oHoo.m I ca.oHNH.mI oH.oHac.mI ov.aHmo.H I ms.ono.aI so.oHom.mI zc A3864,3007 oo.oHsm.oI so.QHoo.mIL em.mem.m I om.ono.mI NH.onm.oI mos: om.mHom.HHI 48.0HHH.mI oo.oHHm.oI oa.stv.omI om.oHHm.eaI oo.QHmo.mI om mg.oHo~.oHI NH.onH.oI so.ono.~I oH.HHHa.maI sm.oHo~.sI Ho.ono.mI use a 8.6Hozxasoo Csao:\eso.sv Asaoz\eso.xv A 8.6Hozxasov Asaoz\aso.xv AoHoz\Hso.xv .mo are Cosomvsoo .mo are Cosomvsoo ucs>aom +Amo.mov~mov +Amo.mommmov mqu>H0m maomsquOC mCoHHm> CH mUvNaa CCm mUnmaa ma CoH +mU mo CoHummeano How mmeHqusv UHEmmeoEHmnu mo COmHHmaEOU .vm mHamB 127 Another important point is that in all cases (with the exception of acetonitrile and methanol solutions) the enthalpies of complexation between Cs+ ion and dibenzo- 27-Crown-9 are more negative than those for the correspond- ing reaction with dibenzo-24—Crown-8. These results may reflect to the increased electrostatic interaction of the Cs+ ion with dibenzo-27-Crown-9 because of increasing number of interaction sites. On the otherlunul.the greater flexibility of dibenzo—27-Crown-9 may allow for better enclosure of the cesium ion. In another word, the con- formational rearrangement of the ligand upon coordination may occur so as to allow for the greatest interaction between the oxygens of the crown ether and the cation. Such rearrangement has been observed for dibenzo-BO-Crown- 10 on complexation with K+ ion as pointed out in Chapter I (page 14). In this case a large deformation of the ligand occurs as it wraps around the potassium ion. It seems reasonable to assume that some wrapping around may occur in the case of complexing of Cs+ ion by dibenzo-27- Crown-9 which would be expected to result in larger electrostatic energies associated with the Cs+ donor atoms than in the case of complexing of Cs+ ion by dibenzo-24- Crown-8. An obvious exception is seen in the case of acetonitrile solutions, probably because it can form a complex in solution with the polyethers, such as that re- ported with 18-Crown-6 (107). We have no explanation for 128 the thermodynamic data of the (DB27C9-Cs)+ system in methanolic solutions. Of course, the role of the ligand solvation on the thermodynamics of complexation cannot be ignored. Since at the present time, the information on the interaction of macrocyclic ligands with the solvent molecules is quite sparse, additional studies on the ligand-solvent interaction are necessary before the thermo- dynamic behavior of macrocyclic complexes in nonaqueous solvents are understood. CHAPTER V CESIUM-133 NMR STUDY OF THE CESIUM ION COMPLEXATION BY CROWN ETHERS AND CRYPTAND-ZZZ IN MIXED NONAQUEOUS SOLVENTS 129 A. Introduction While the complexation of macrocyclic ligands with the alkali and alkaline earth cations has been extensively investigated in a wide variety of pure solvents, the com- plexing ability of the crowns and cryptands in mixed non- aqueous solvents has not been reported previously. Pre- vious measurements of the binding constants of macrocyclic complexes in mixed solvents have been restricted to water/ methanol systems. It was of interest to us to perform some thermo- dynamic measurements in mixed nonaqueous solvents in order to see how the nature of the medium affects the thermo- dynamic stability of crown and cryptate complexes. This chapter reports a study of the Cs+ ion com- plexation by macrocyclic ligands C222, DB24C8, and DB27C9 133Cs in various mixed nonaqueous solvents. Once again NMR was used as the experimental technique. B. Complexation of the Cs+ ion by C222 Complexation of the cesium ion with cryptand-222 was studied in AC/DMSO, AN/DMSO, PC/DMSO and PC/DMF binary systems at the probe temperature (32 i 1°C). Cesium-133 chemical shifts were determined as a function of C222/Cs+ mole ratio. The results are listed in Tables 35-38. In 130 131 oo.ao ~m.s mm.am mo.e mo.mm No.4 so.oo mm.e mo.~m He.m ~m.om om.m He.mm ma.m om.am mm.a ma.ma mo.~ Hm.ms 4m.~ mo.om as.m mm.mm ~o.m mm.ss mm.~ e~.oa oH.m om.ma 8H.H mm.om mo.~ mm.ma 4m.H me.ms ma.a os.He HN.H om.aa OH.N 40.4a mm.a Ha.ao NN.H mm.oo om.o He.mc so.a ma.om oo.H ca.4o mm.o om.ao Ha.o -.ma oo.H mo.mo om.o mo.Ho em.o mm.mo mm.o Ha.mo mo.o om.mo mm.o am.mm om.o Hm.eo Hm.o om.oo -.o om.oo mm.o mo.mm oo.o No.ao oo.o mm.mo oo.o ma.mo oo.o Aaaavc +mU\a AEaavm +mU\a Aaaavm +mU\a_ Aeaavm +mU\a om mm~\omze ems om mmH\omzo mmm om mm\0mzo ems Omzo muse .oomm um mamummm om\omzo sH NNNU mo woCmmwua 020 CH AZmoo.ov ZUmmU How mama HUHCm HmoHEMCUIoHumm 0H0: .mm mHamB 132 mo.aoa ma.s oo.mma om.s ac.ama oe.s mm.mea om.e oe.so ma.m mo.oma oe.m He.oaa m4.m oo~.oa 84.4 Hm.ema mm.m am.mm ve.~ Ha.omH om.m Ho.oaa oa.~ mm.ama mm.m mo.m~a mm.~ mm.om ma.m em.cma No.~ mm.maa mm.m mm.oma am.~ HN.HHH ma.~ mo.me sm.a Hm.maa oo.a om.coH oo.a so.oeH mm.a mm.aHH sm.a ma.4a ~m.a HH.mmH ea.a mm.omH om.H mm.mHH m~.H oa.moH me.a mm.oa NH.H ~m.aea oa.o mm.oMH ao.a m~.ooa co.H Ho.Ho NH.H Hm.mo oo.H ea.maa ma.o ma.maa mm.o mH.Ho ms.o mm.mm mm.o ms.ao oo.o sc.oo om.o mo.om om.o me.eo mm.o mm.mm He.o mm.oo mm.o mm.m~ 4~.o Hm.Hm o~.o om.mm om.o oa.mm e~.o 6H.mm om.o mm.H~I oo.o Hm.mI oo.o m~.s oo.o Ho.o~ oo.o mH.He oo.o AEaavc +mU\a Aeaavm +mU\a AEaavw +mU\a Asaavc +mU\a Aeaavm +mU\a om muss om mmoxomzo mm 04 mmm\omzo mma. om mms\0mzo mmm om mom\omzo mom AcmaCHuCoov mm OHQmB 133 mm.mo~ oe.e ma.sma oo.m mo.mma oo.m oa.mom -.m mo.HoH oa.e am.m~a oo.m Hm.4ma e~.m Ho.emH oo.~ mo.mom oa.~ mm.eo me.m mm.mma Hm.m om.oaa oa.m me.HmH mo.~ Ha.mo~ ~m.~ ee.mo oa.m oo.mHH ca.~ 4H.H4H oH.~ oo.mHH ~8.H oa.mom om.H Hm.oe se.~ oo.HHH -.~ Ha.oma mo.a Ho.ooH om.H ao.eo~ mm.H o~.mm om.a om.ma ms.a om.oma om.H oo.mma mo.a m~.aoa oo.H mm.am ov.a Hm.oo mH.H mo.soa oo.a oH.HmH mm.o mo.moa om.o oH.oH eo.H 4H.om ma.o ss.~o oo.o m~.oHH em.o no.8HH om.o om.ao mm.o so.~e He.o No.48 os.o mo.~m ~m.o os.~o -.o em.ao oo.o ma.mm oo.o m~.me oo.o am.aa oo.o oa.mm oo.o Afiamve +mU\a Asaavc +mU\a Asaavm +mU\a AEaavm +mU\a AeamWo +mU\a omze mom\z< mom omzo moo\z< mos omze mos\z< moo Oman o~\z« wow 24 when .0 mm us uewumam omze\zm ca zomso smoo.o cue; stCEoo ammo mo scsum cause sacs .om sense 134 mH.me mo.v va.vm mm.m mH.mHH hm.¢ mh.HvH om.v mo.mmH mv.m Hm.om om.m mn.moH vo.m mo.mMH wH.m mH.HmH om.m mm.mm mm.~ Hm.mm mg.m m>.vNH mo.m mm.omH mm.H m~.Hm mo.m mN.Hm mm.H mh.vHH mm.H om.mMH Hm.H Hm.>h Ho.H mm.mm mv.H mm.hoH mm.H mm.mmH 0H.H mm.mm FN.H mm.¢> vo.H om.Hm wH.H om.HHH om.o om.mm Hm.o mo.mo mm.o mm.hm mm.o om.om mm.o m>.oo m~.o mm.vm mm.o mm.om mv.o No.mm m¢.o vw.>m oo.o hm.mv oo.o om.mm oo.o No.h oo.o Afiaavm +mU\a Aeaavo +mU\a Asaavm +mU\a Afiaavm +mU\a omzo mom\os mom omso moo\om mos omzo moexom woo omza mo~\om mom .Usmm um moustHz Hsmch 0mzo\oo CH zommo Cmoo.o ans xwaeaoo mmmo mo smsum oHusm 0H0: .hm mHQmB 135 oo.mma No.4 mm.m~H ms.c m~.moa mv.4 oa.mo mc.s mm.~ma 80.4 ~4.Hma ~m.o sm.o0H so.s HH.~m mm.m mm.mma mo.m me.~HH mm.m mo.mo mm.m NH.mm o~.m oo.gaa se.~ Hm.moa mo.~ mm.em om.m eo.om om.m NH.ooH so.m oo.mo ea.m mm.os om.~ mo.Ho mm.a Ho.om mo.H ca.em mm.H oo.oc om.a ov.om ae.a om.mm mm.H oo.mo mm.a om.om om.H Hm.mm mo.H mm.Ho ~o.o mm.om mm.o mo.me oo.H Hm.om om.o mo.m4 so.o mm.~m mm.o om.om mono mo.m~ mm.o Ho.mH o~.o Hm.HH em.o m~.4a Hm.o om.m mm.o mg.mI oo.o ma.mI oo.o Ho.mI oo.o om.HI oo.o Asaavm +mU\a Asaavm +wU\a Aeaavm +mU\a Asaavw +mU\a om mom\mzo mom on mo~\mze wow 08 moa\mze mom C28 seam .oomm us msmummm om\mzo CH Nmmo Co consumes 0:» CH Azmoo.ov zommo sow memo uaHCm HmoHECCo I oHumm 0H0: .mm canoe 136 44.4ma 44.4 oo.4ma oo.m ao.~ma 44.4 4m.asa 44.4 ma.osa 44.4 -.mma Hm.4 mo.mma H4.m ~4.ama om.m Ne.maa oo.m 4o.oca mo.m ~H.mma ma.4 mm.~ma mo.~ mm.oma 4~.m os.mea mm.m om.aoa mm.m Ho.m4a H4.m om.~ma m4.~ ~m.maa 44.~ m4.maa 4m.~ -.moa mm.m H~.m4H He.~ Hm.osa 4m.H 4m.aaa oa.a aa.osa om.~ Hm.ama NH.~ Hm.mma 4~.m H4.oca o4.a 4o.moa om.a o4.4oa 4m.a mo.mma om.a 44.4NH m4.H ma.moa mm.a ma.m4a oo.H am.m4a -.H ma.a4a om.a om.aoa mm.a am.~ma mo.o mm.maa 4e.o ~4.moH ma.o mm.oHH ma.o sa.mm ma.o H4.ooa mo.o oa.oo m4.o 44.44 oo.o Ho.oo 4m.o o~.mm em.o 84.4H ~.o 4o.mm mm.o om.m~ o~.o oa.a~ o~.o m4.mH 4~.o ma.smI oo.o om.amI oo.o ma.4mI oo.o oo.oHI oo.o 4a.~HI oo.o Cameos +uo\e Cameos +mo\e remove mo\q Cameos mo\e Cameos moxe om muse om moo\mzo moa om mom\mzo mom oo moa\mze mom on mom\mzo mom AmeCHuCoov mm wHamB 137 all cases only one resonance of the metal ion was observed,’ suggesting that the exchange between the two sites (free Cs+ ion and complexed-Cs+ ion) is fast compared to the NMR time scale. Graphical representations of the results are shown in Figures 35-39. It is immediately obvious that the nature of the medium plays an important role in the com- plexation process. For example, as the mole fraction of acetone, acetonitrile, and propylene carbonate in AC/DMSO, AN/DMSO, and PC/DMSO systems is increased, the plots show more pronounced curvature which is evidence for the forma- tion of a stronger complex. Similar behavior is observed in the case of PC/DMF system as the fraction of the propylene carbonate increases. The values of the formation constants for the cesium-cryptate complex, calculated from the variation of the metal ion chemical shift with the C222/Cs+ mole ratio, are given in Tables 39-41. The data collected in these tables again show that the values of the stability con- stant (Kf) of the Cs+-cryptate increases as the concentra- tion of dimethylsulfoxide and dimethylformamide is lowered. The behavior observed for these systems seems reasonable because solvents with a large donor number, such as dimethylsulfoxide (D-N = 29.8) and dimethylformamide (DoN = 26.6) have fairly large solvating ability for the cesium ion, but solvents with a medium Gutmann donicity, like acetone (D-N = 17.0), acetonitrile (D-N = 14.1), and 138 I90 " fl.“ _*_ I) C) \‘ n '1 I30 - ‘\ ”s / ° D. 70" I" 2’. o. J” V c 4' ‘0 * Pure AC 0 95H% "‘3 o 357 Ac 75% Ac [3 50% AC ...50— l I I l L V 0-00 1.00 2.00 3.00 4.00 5.00 [0222]/[0s+] Figure 35. Cesium-133 chemical shifts vs. (C222)/Cs+) mole ratio in DMSO/AC binary mixtures. 139 940t- O . b C t3€5lD-- , Id ./ / / 78.0— " // 1/ I /' E _. / ‘ Et'71313 .2 . V ' . .' a —> 75%0M50+25%AC ao . s —> 85% DMSO + I5°/.AC 52.0, c -+ 95%DMSO + 5% AC 0! —+ Pure DMSO 540- ¢1€ii)-- I I I I C1!) LC) 21() 311) 41!) £1!) [0222]/[0s+] Figure 36. Cesium-133 chemical shifts vs. (C222)/(Cs+) mole ratio in DMSO/AC binary system. 140 22|55—' l6355- l£555- T~<=3SNE\ 3555 I a —> Pure AN 0 —> 80% AN + 20% DMSO 0 —> 60% AN + 40% DMSO 0 —> 40% AN + 60% DMSO s —> 20% AN + 80% DMSO I I | LC) EL!) 25!) ll!) [02 22] /[Cs”] Figure 37. Cesium-133 chemical shifts vs. (C222)/(Cs+) mole ratio in AN/DMSO binary solutions. 51!) 141 £2()C) I I I I O 0 I60— — C I20 — d- e ’E‘ 80 -— — 2. r/ v 7'! a —> Pure PC ‘0 40 b —» 80% PC + 20% DMSO _ V c —» 60%PC + 40%0M50 d —» 40% PC + 60%DMSO O s —» 20%P0 + 80%0M30' —30 I l l I 0.0 I .0 2.0 3.0 4.0 5.0 [0222]/[0s+] Figure 38. Cesium-133 chemical shifts as a function of (C222)/(Cs+) mole ratio in PC/DMSO binary mixtures. 142 ISO — 0 - 8 d 0 I50 - f 9 h ,1 HO"- E k a. 3 ‘n .7() —_- l2 4. ' a—-> Pure PC 0 30 7' s—» 909'. PC d—> 70% PC -IO e—> 50% PC I; f—> 30% PC I g—> 20% PC h—> IO°/s PC -50_ k—-> Pure DMF I I I I J 0.00 I00 2.00 3.0 4.00 500 [0222] /[Cs+] Figure 39. Cesium-133 chemical shifts vs. (C222)/(Cs+) mole ratio in DMF/PC binary solutions. 143 Table 39. Log of Formation Constants and Limiting Chemical Shifts for (C222-Cs)+ Complex in AC and DMSO and Their Mixtures at 32 :_l°C Solvent Log Kf Limiting Chemical Shift Pure AC 3.75 i 0.02 191.1 : 0.1 95% AC + 5% DMSO 3.28 i 0.01 187.2 1 0.2 85% AC + 15% DMSO 2.80 i 0.01 176.8 i 0.3 75% AC + 25% DMSO 2.41 i 0.02 156.7 : 1.0 50% AC + 50% DMSO 1.80 i 0.01 155.7 i 1.5 25% AC + 75% DMSO 1.38 i 0.04 163.5 i 7.7 15% AC + 85% DMSO 1.31 i 0.03 156.1 i 3.8 5% AC + 95% DMSO 1.25 i 0.04 143.7 : 6.0 Pure DMSO 1.19 i 0.02 154.2 i 3.4 144 Table 40. Log of Stability Constants and Limiting Chemical Shifts for (C222-Cs)+ Complex in Binary Mixed Solvents (at 32 : 1°C) Solvent Log Kf Limiting Chemical Shift AN/DMSO System Pure DMSO 1.19 i 0.02 154.0 : 3.0 20% AN/80% DMSO 1.68 :_0.02 148.0 : 3.0 40% AN/60% DMSO 2.05 i 0.03 167.4 1 3.0 60% AN/40% DMSO 2.52 i 0.03 179.4 : 1.6 80% AN/20% DMSO 3.27 i 0.02 193.9 1 0.02 Pure AN 4.71 i 0.06 206.0 i_0.3 PC/DMSO System 20% PC/80% DMSO 1.55 i 0.04 155.6 i 5.6 40% PC/60% DMSO 1.94 i 0.01 159.4 1 1.2 60% PC/40% DMSO 2.36 i 0.02 170.8 : 1.2 80% PC/20% DMSO 3.09 i 0.07 173.7 : 1.3 Pure PC 3.90 i 0.01 186.0 1 0.1 145 Table 41. Log of Stability Constants and Limiting Chemical Shifts for (C222-Cs)+ Complex in PC and DMF and Their Mixture at 32 i.1°C' Solvent Log Kf Limiting Chemical Shift Pure PC 3.90 i 0.01 186.0 1 0.1 90% PC + 10% DMF 3.65 i 0.02 184.8 1 0.2 80% PC + 20% DMF 3.45 i 0.03 181.1 i 0.4 70% PC + 30% DMF 3.19 i 0.01 178.0 : 0.3 50% PC + 50% DMF 2.80 i 0.03 169.6 i 1.2 30% PC + 70% DMF 2.45 i 0.01 163.8 : 1.0 20% PC + 80% DMF 2.29 i 0.01 159.8 : 0.8 10% PC + 90% DMF 2.14 i 0.02 154.8 : 1.9 Pure DMF 2.00 i 0.02 150.1 : 3.0 146 propylene carbonate (D°N = 15.1) have relatively poorer solvating prOperties towards the cesium ion. A metal ion which is totally surrounded by the ionophore (ligand) should exhibit a chemical shift in— dependent of the solvent. The chemical shifts of the cesium ion bond to the cryptand-222 in AC/DMSO, AN/DMSO, PC/DMSO, and PC/DMF are listed in Tables 39-41. It is interesting to note that, in all cases, the limiting chemical shifts for the complexed-cesium ion are depen- dent on the solvent compositions, suggesting that the cesium ion is only partially insulated from the solvent and that the complex exists in the exclusive configura- tion (11). C. Complex formation by Cs+ ion with DB24C8 in DMSO-AC, and PY-MeOH binary systems In the present work, the influence of the solvent on the stability of the (DB24C8-Cs)+ complex was investi- gated. For this purpose, the chemical shift of cesium-133 resonance was measured as a function of DB24C8/Cs+ mole ratios in dimethylsulfoxide-acetone and pyridine-methanol solutions at 32 i 1°C. The data obtained from these studies are given in Tables 42 and 43. Figures 40-42 show the variation of the 133Cs chemical shifts versus DBZ4C8/Cs+ mole ratios at different composition of the solvents. Investigation of these figures shows that in dimethylsulfoxide-acetone 147 44.44 44.4 44.4 44.4 44.4I 44.4 44.44I 44.4 44.44 44.4 44.4 44.4 44.4I 44.4 44.44I 44.4 44.44 44.4 44.4 44.4 44.4I 44.4 44.44I 44.4 44.42 44.4 44.4 44.4 44.4I 44.4 44.44I 44.4 44.44 44.4 44.4 44.4 444.4I 44.4 44.44I 44.4 24.44 44.4 44.42 44.4 44.4 44.4 44.44I 44.4 44.44 44.4 44.44 44.4 44.4 44.4 44.44I 44.4 44.44 44.4 44.44 44.4 44.4 44.4 44.44I 44.4 44.44 44.4 44.44 44.4 44.4 44.4 44.44I 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44I 44.4 282224 +8o\2 228224 +8o\2 282824 +8o\2 288424 +8o\2 0422 444\oa 444 0422 4om\om 444 0422 444\02 444 02 00.44 08 8422242 04824 428 0422 828 02 24 404482 40 80288848 822 24 22444.44 20480 404 8882 84424 48848820I04882 8402 .44 84288 148 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 44.44 44.4 224444 +40\4 422244 H40\.4 254444 +40\4 0420 8444 0420 444\0¢ 444 0420 444\0< 444 24822442084 44 84244 149 44.44- 44.4 44.44- 44.4 44.44- 44.4 44.44: 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.44: 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.44: 44.4 44.44: 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.44: 44.4 44.44: 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.44: 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.4 n 44.4 44.44: 44.4 44.44- 44.4 44.44: 44.4 44.4 u 44.4 44.44: 44.4 44.44: 44.4 44.44- 44.4 44.4 44.4 45mm44 +4o\4 Aemmvo +mU\A AEmmvo +mo\4 459044 +mo\q 44 444\4082 444 44 444\4082 444 44 444\4082 444 44 444\2082 444 0044 44 4284444 44\2082 24 404440 40 80284844 824 24 42444.44 20440 404 4440 44424 44040820u04444 8402 .44 8424B 150 mm.mml mm.m 00.mm| mm.m m0.4mt 4m.m 4N.0ml 04.m mm.mml N4.m 44.le 0N.m 40.4mn 4m.m 4m.0m| 04.4 mm.mm| 00.N 44.4ml 0m.~ 40.4ml m4.m 0m.0m| 4m.~ 00.mm| 44.4 mm.mmu 40.4 40.4mu 44.4 40.4w: 40.4 mm.mm| 44.4 mm.mmu 0m.4 40.4ml 4m.4 m0.mm| 04.4 m4.mm| No.0 mm.mm: N4.4 M4.4mu 04.0 44.4Nn 40.0 44.44: 40.0 mm.mm| 40.0 44.44: 04.0 04.4w: m4 m4.4m| 44.0 mm.mm| 44.0 04.44: 4m.0 44.4w: 44.0 44.mm| 0N.0 44.mmn 44.0 40.nml mm.0 04.0w: 04.0 mm.mml 00.0 40.mm| 00.0 04.mml 00.0 44.4w: 00.0 4Emm04 +mo\4 AEQQ04 +mo\q Aammvo +mU\A AEvae +mU\A 44 444\2082 444 44 444\4082 444 44 444\4082 444 44 444\2082 444 Awwscwucoov mv mHQMB 151 04.mml mm.m 44.44! 40.4 mm.mml 4m.m 04.mmn 4m.~ 4m.4m- No.4 44.mmu mm.~ 44.mmu No.4 4m.4ms 44.4 04.mmu 44.4 mn.mmu 4m.4 44.4m- 4m.4 44.mmu 44.4 4m.om- mm.4 40.mm- 44.4 4m.mmu 44.4 44.4m- 40.4 mm.4mn 44.0 44.4mu 04.0 m4.4m- 44.0 44.4m- mo.0 m0.mmu mo.0 4m.04u 4m.0 44.4m: m4.0 4m.4mu 4m.0 m4.m4- 04.0 44.0m- 44.0 44.4mu 44.0 4m.m4- 00.0 04.04- 00.0 m0.4mn 00.0 454400 +m0\q 454400 +mo\q 424400 +m0\q 4m 44\momz 4mm 44 4m4\4omz 4mm 44 4m~\4omz 4mm 40w904ucoov 44 @4949 152 70‘ 64 56 8(ppm) 4o.— 24— o — Pure DMSO b —— l5°/oAC /85°/oDMSOE C —- 3OO/0AC/ 700/0 DMSO l l l 0.0 Figure 40. LG 2.0 3.0 4.0 [DB 24cs] /[Cs+] + Cesium-133 chemical shifts vs. (DB24C8)/Cs ) mole ratio in AC/DMSO systems. 5.0 153 36‘ a —> 50%AC / 50% DMSO '3 -»*7<)94:l\C2IESCT9QIJWVESC) c —> 85% AC / l 5% DMSO d —>Pure AC 28 20 AIZ E o o. o. 00 4.. b ..4— xi. V C) "12E3'_' 43—- (j l I l I 0.0 LO 2.0 3.0 4.0 5.0 [0824cs]/[Cs+] Figure 41. Cesium-133 chemical shifts vs. (DBZ4C8)/(Cs+) in’AC/DMSO systems. 154 systems, as the concentration of the dimethylsulfoxide increases the plots show less curvature which implies the formation of a weaker complex at high dimethylsulfoxide concentrations. It is interesting to note that, the addition of the ligand to a solution of Cs+ ion in DMSO/AC mixtures always results in an upfield shift. In the PY/MeOH systems, however, the behavior of the chemical shift is quite un- expected. As seen in Figure 42, the variation in the cesium-133 chemical shifts as a function of the mole ratios is not significant in about 50-60 Vol. % of methanol con- tent in PY/MeOH systems. It seems that, the electronic environment of the Cs+ ion does not change upon complexa- tion in this range of the solvent composition. The stability constants of the (DB24C8°Cs)+ com- plex in DMSO/AC and PY/MeOH systems are listed in Tables 44 and 45. As shown in Table 44 the stability constants of the complex increases with increasing the acetone con- centration in the DMSO/AC mixtures. In the case of PY/MeOH systems, however, there does not seem to be a simple relation between the stability of the complex and the solvent composition. Discussion The variation of the formation constants of the (C222-Cs)+ complex as a function of solvent composition in acetone-dimethylsulfoxide, acetonitrile-dimethysulfoxide, Figure 42. Cesium-133 chemical shifts as a function of (DBZ4C8)/(Cs+) mole ratio in methanol- pyridine binary solutions. 155 o —— l5% MeOH / 85%PY ... b — 30% " / 70% " C — 40% u / 600/0 u d - 45% " / 55% " -7 +— e -— 50% " ’ / 50% " f —— 60% " / 40% " 9 — 6570 u / 350/0 u -| 3 h — 70% " / 30% i —— 75% " / 25% " j — 85% " / - l5% " “20 k — 93% " / 7% " E a g '0 -26 60 b C I g -32 W_ % 1' k ~38 ' -44 I J J l 0.0 l.0 2.0 3.0 4.0 [0324cs]/[Cs+] 5.0 156 Table 44. Formation Constants of (DB24C8-Cs)+ Complex in DMSO-AC Mixed Systems (at 32 : 1°C) Solvent Log Kf Pure AC 3.77 i 0.06 85% AC + 15% DMSO 2.91 i 0.02 70% AC + 30% DMSO 2.38 i 0.02 50% AC + 50% DMSO 1.94 i 0.01 30% AC + 70% DMSO 1.73 i 0.03 15% AC + 85% DMSO 1.44 i 0.07 Pure DMSO 1.44 l+ O O .p. 157 Table 45. Formation Constants of (DBZ4C8-Cs)+ Complex in MeOH-PY Binary System (at 32 i 1°C) Solvent Log Kf Pure MeOHa 3.65 i 0.05 93% MeOH + 7% PY 3.32 i 0.08 85% MeOH + 15% PY 3.42 i 0.11 75% MeOH + 25% FY 3.47 i 0.10 70% MeOH + 30% FY 3.45 i 0.10 65% MeOH + 35% PY 3.71 i 0.24 50% MeOH + 50% FY 3.31 i 0.12 45% MeOH + 55% PY 3.52 i 0.09 40% MeOH + 60% FY 3.59 i 0.11 30% MeOH + 70% PY 3.69 i 0.07 15% MeOH + 85% PY 3.94 i 0.06 Pure PYa 4.00 i 08 aReference 70. 158 prOpylene carbonate-dimethylsulfoxide, and propylene carbonate-dimethylformamide solutions are shown in Figures 43 and 44. Investigation of these figures shows that the complexation process in mixed nonaqueous solvents is quite sensitive to the solvent composition. It is interesting to note that in DMSO/AC solu- tions the variation of the stability constants of the (C222-Cs)+ complex versus the solvent composition is not significant at low acetone concentrations (between 0 to ~ 15 Vol. % of acetone), suggesting that the metal ion, and possibly the ligand, are solvated only by the dimethyl- sulfoxide molecules. At higher contents of the acetone (> ~ 15 Vol. % of acetone), however, there is a substantial increase in the formation constants of the complex. It seems reasonable to assume that the dimethylsulfoxide molecules in the first solvation shell of the metal ion (and probably of the ligand) are replaced gradually by the acetone molecules upon addition of more acetone to the system. ~Since acetone is a much weaker solvating solvent than dimethylsulfoxide, the stability of the complex in inhanced by diminished complexation of the solvent for the metal ion. Another important point which can be deduced from the comparison of the plots in Figures 43 and 44 is that the curvature of the plots of log K against the f composition of the binary solutions is more pronounced in the cases of acetone-dimethylsulfoxide and acetonitrile- 159 l I l l O 20 4O 60 80 IOO V% of Solvent 3 in (DMSO/S System) Figure 43. Variation of the stability constant of (C222-Cs)+ complex with solvent composition in different binary solutions. 160 4:. l 4. I09 Kf Of (szz' CS)—"’ N w l 1 I 1 I 41 I I I J Figure 44. 20 4O 60 80 IOO v% of PC in (PC+ DMF) Variation of the formation constant of (C222-Cs)+ complex with PC content in PC/DMF binary mixtures. 161 dimethylsulfoxide solutions than in the cases of propylene carbonate-dimethylsulfoxide and propylene carbonate- dimethylformamide systems. This is probably due to the differences in the cation and ligand solvation in these systems caused by changing of the medium dielectric con- stant, changing of the donor ability of the solvents, and variation of the solvents structure resulting from changing in composition of the mixed solutions. Figure 45 shows the stability constants of the (DBZ4C8-Cs)+ complex as a function of the pyridine and acetone concentrations in the pyridine-methanol and acetone- dimethylsulfoxide binary systems. A regular decrease in the formation constant of the complex in AC/DMSO mixtures is observed as the concentration of DMSO increases. It is interesting to note that, while the stability constant of the (DBZ4C8-Cs)+ complex varies monotonically with the solvent composition in AC/DMSO binary solutions, this trend is not followed in PY/MeOH systems. With in- creasing the concentration of the pyridine, the formation constant of the (DBZ4C8-Cs)+ complex decreases initially to reaCh a minimum (at ~ 10-20 Vol. % of pyridine) then increases to a maximum (at ~ 25 Vol. % of pyridine) and decreases again to about 50 Vol. % of pyridine. This behavior may reflect changes occurring in the structure of the solvent mixture or in the solvation properties of the cyclic polyether and the cation at these solvent 162 4 .0 PY MeOH ‘ AC L09 Kf l0 0 I DMSO I.O —- 0.0 I l I J l I J J l 0 IO 20 3O 4O 50 60 70 80 90 I00 Composition of Binary Solution (V0l.%) Figure 45. Variation of the stability constant of (DBZ4C8-Cs)+ complex with the solvent composi- tion in MeOH/PY and DMSO/AC binary solutions. 163 compositions. Some structural changes probably occur in the structure of the solvents when they mix with one an- other. These structural changes may result in changing the interaction of those solvents with the solutes com- pared to the pure solvents. The estimated formation constants (derived from the plots in Figures 43 and 45 of the (C222-Cs)+ and (DBZ4C8-Cs)+ complexes and the inverse of the dielectric constants of the medium in various compositions of acetone- dimethylsulfoxide systems are listed in Table 46. Figure 46 shows the stability constants of these complexes as a function of the reciprocal of dielectric constant of the medium. As shown in this figure, the plots of log Kf against the inverse of the dielectric constant are slightly concave downward. Such a curvature could be explained in terms of specific solvent effects which may result from the chemical nature of the solvent in the mixed solutions. The existence of a linear relation between the logarithm of the stability constants of the various com- plexes and the reciprocal of the dielectric constant of the solvents has been reported by several researchers (123—126). For example, Tur'yan et al (123, 127, 128) showed that for the complexes of lead chloride and thiocyanate there is a linear relation between pK and % (D = dielectric const.of the medium) for aqueous alcoholic solutions. It should be noted that, this relation may 164 Table 46. Stimated Formation Constants of the (C222°Cs)+ and (DB24C8-Cs)+ Complexes in Various Composi- tions of AC/DMSO Binary Solutions Log K f Solvent D* % (C222-Cs)+ (DB24C8-Cs)+ Pure AC 86.4% AC/l3.6% DMSO 20.7 0.048 3.75 3.77 83.4% AC/l6.6% DMSO 24.0 0.042 2.90 3.05 75% AC/25% DMSO 27.0 0.037 2.40 2.55 69.5% AC/20.5% DMSO 28.5 0.035 2.20 2.35 58.3% AC/4l.7% DMSO 31.6 0.032 1.92 2.07 37.6% AC/62.4% DMSO 37.3 0.027 1.55 1.75 27.5% AC/72.5% DMSO 39.9 0.025 1.45 1.65 Pure DMSO 46.6 0.021 1.19 1.44 * D = Dielectric Constant (Reference 144). 165 4.0 '— 3.0 — 0.... X as 2 0 -- O .J b I.O — 0.0 I I 1 I 1 I 1 I LO 2.0 3.0 4.0 5.0 J5 x102 Figure 46. Variation of the stability constants of (a) (DBZ4C8-Cs)+ and (b) (C222-Cs)+ complexes with the reciprocal of the dielectric constant of the medium in AC/DMSO binary solutions. 166 only hold in the case of simple ion-ion reactions in chemically similar solvents, with no specific ion solvent interactions. However, this may not be true for the molecular-ion reactions. Since the relation between log K and % is not linear in the case of (C222-Cs)+ and (DBZ4C8-Cs)+ systems in AC/DMSO binary solutions, it can therefore be concluded that the stability of these complexes is a function of chemical nature of the solvent as well as of its dielectric constant. Knowledge of the molecular state of the solvent and its nature in a binary mixed solvent is important in understanding the effect of the solvent on the complexation process. D. Thermodynamic Parameters for Complexing of Cs+ ion by DB27C9 and DB24C8 in DMF/AN Binary Solutions. In the present investigation, the interactions be- tween large macrocyclic crown ethers, dibenzo-27-crown-9 and dibenzo-24-crown-8 with cesium ion were studied in dimethylformamide/acetonitrile binary mixtures at various temperatures. At each temperature, the chemical shifts of the cesium—133 were monitored as a function of DB27C9/Cs+ or DB24C8/Cs+ mole ratios. In all cases, only one popula- tion averaged NMR signal was observed irrespective of the Ligand/Cs+ mole ratio. 167 The cesium-133 chemical shifts measured at different temperatures for various systems are given in Tables 47-51 and the plots of the 133Cs chemical shifts as a function of Ligand/Cs+ mole ratios are represented in Figures 47-51. These plots show that the equilibrium between the Cs+ ion and dibenzo-27-crown-9 and dibenzo-24-crown-8 in DMF/AN binary systems are temperature dependent. As the tem- perature decreases the plots show more pronounced curvature which is an evidence for the existence of exothermic re- actions between the cesium ion and these ligands. The formation constants of (DB27C9-Cs)+ and (DBZ4C8-Cs)+ complexes for the various compositions of DMF/AN system at different temperatures are listed in Tables 52 and 53. A monotonic increase in the stability constants of these complexes is observed as the tempera- ture decreases. Plots of ln Kf versus % for the data in Tables 52 and 53 are shown in Figure 52, and the correspond- ing thermodynamic parameters (0H2, ASé and AG° values) are given in Table 54. Thermodynamic data show that in all cases the complexes are enthalpy stabilized but entropy destabilized and the values of 0H2, ASE, and AGE are dependent on the solvent compositions. Discussion An evaluation of the changes in H2 and SE as a function of the change in the composition of the mixed solvent is interesting, because these values may be 04.441 44.441 44.441 44.441 4m.041 40.441 04.4 168 44.441 44.441 44.441 44.441 44.441 40.041 40.4 44.441 44.441 44.441 44.441 44.441 44.441 44.4 44.441 44.441 44.441 44.441 44.441 40.441 44.4 44.4 1 44.041 44.441 44.441 44.441 40.441 44.4 44.4 1 44.4 1 44.4 1 44.041 40.441 40.441 40.4 44.4 44.4 04.0 1 44.4 44.4 44.4 44.0 00.4 44.4 44.4 1 44.4 44.4 04.4 44.0 44.4 44.4 40.4 1 40.0 44.0 44.4 44.0 40.4 04.0 44.4 44.4 44.4 44.4 00.0 04 04 04 04 0 041 wusumummfima +mU\.4 mwusumummsma m50440> um 429 404 + 24 404 :4 42400.0V zommo £043 004440 «0 :o4umxw4asoo How mama 044nm 4404swno 1 04444 0402 .44 440mg 169 44.441 04.441 44.441 44.441 44.041 44.041 44.4 44.441 44.441 44.441 44.441 44.441 44.041 44.4 44.441 44.441 44.441 44.441 44.441 44.441 44.4 40.441 44.441 44.441 44.441 44.441 44.441 44.4 44.041 44.441 44.441 40.441 44.441 44.441 04.4 44.4 1 44.4 1 04.041 44.441 44.441 40.441 40.4 04.4 1 44.4 1 44.4 1 44.4 1 44.4 1 44.4 1 04.0 44.4 1 44.4 1 44.0 1 44.0 1 44.0 04.0 44.0 44.0 44.4 44.4 04.4 04.4 . 44.4 44.0 44.4 44.4 40.4 44.4 44.04 44.44 00.0 04 04 04 0 041 041 +mU\4 Do musumquEmB .mmHDumummEmB pamHGMMHQ um MED 404 + Z< 404 :4 42400.0v zommu £443 404440 40 :04umxm4QEou MOM mumo uw4nm 4mo4Em£U 1 04umm $402 .44 m4nme 44.441 44.441 44.441 44.441 44.441 44.441 44.4 170 44.441 44.441 44.441 40.441 44.441 44.441 44.4 04.441 04.441 44.441 44.441 44.441 44.441 44.4 44.441 40.441 04.441 44.441 44.441 44.441 44.4 04.4 1 40.4 1 44.041 44.441 44.441 44.441 44.4 44.4 1 44.4 1 44.4 1 44.4 1 04.4 1 44.041 44.0 44.4 1 44.4 44.4 44.4 44.4 44.4 44.0 44.0 1 44.0 44.0 44.0 44.4 40.4 44.0 44.4 44.4 44.4 44.4 44.4 44.4 44.0 44.44 44.44 44.44 44.44 44.44 44.44 00.0 04 04 04 MN 04 0 +mU\4 Uo musumquEwa mmusumumewB m504um> um min 404 + 24 404 :4 zummu zmoo.o nu43 mmwaanu 4044mm now mama “MHsm HmOHEmsu 1 OHumm w4oz .44 magma 171 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.441 44.041 40.441 44.441 44.441 44.441 44.441 44.4 1 44.4 1 44.4 1 44.041 44.441 44.441 44.4 1 44.4 1 44.4 1 44.4 1 04.4 1 44.041 44.4 1 04.4 1 44.4 1 40.4 1 44.4 1 44.4 1 44.4 1 44.4 1 44.4 1 44.4 1 44.4 1 40.4 1 40.0 1 44.0 44.0 40.4 44.4 44.4 44.4 44.4 44.4 44.4 44.4 04.4 04 04 04 04 0 041 Do musumummfiwe 44.4 40.4 40.4 44.4 04.4 40.4 44.0 44.0 44.0 00.0 m + U\4 .mmusumu0450e m004um> pm 420 WO4 + 24 wom 04 Asmoo.ov £043 4044mm 40 coflumxwfimsoo H04 00mm unfinm 4m0450no 1 0400m 0402 .04 m4DMB 172 44.041 00.041 40.041 00.041 40.441 44.041 00.4 00.441 00.041 40.041 00.041 00.041 00.041 04.0 44.441 44.041 m4.m41 40.041 04.041 00.041 04.4 44.041 00.441 40.441 00.441 mm.m41 44.041 44.4 04.041 44.441 40.441 40.441 44.441 00.441 40.4 44.0 1 04.4 1 40.0 1 40.0 1 04.041 40.441 04.4 40.0 1 40.0 1 44.0 1 00.4 1 00.0 1 00.0 1 00.0 44.4 1 40.4 1 00.4 1 04.4 1 04.4 1 04.0 1 04.0 00.0 04.0 m4.m 00.4 om.m 44.0 44.0 04.0 44.0 00.0 00.04 04.04 00.04 04.0 40.04 40.44 04.44 00.04 00.44 04.04 00.0 om 04 00 04 04 0 U0 04940404509 +mU\4 000000000500 000000040 00 020 000 + Z0 004 04 42000.00 20000 0043 x040000 004400 00 0000 00400 40042000 1 04000 0402 04009 173 (D “g\ _4_% E-e— 3 00 -12-— ' -l6 — . 40°C 30 20 -20—- '3 ‘IO 1 J 1 I o 1.0 2.0 3.0 4.0 5.0 [08 27 <29] /[Cs +] Figure 47. Cesium-133 chemical shifts vs. (DBZ7C9)/(Cs+) mole ratio in 10% AN/90% DMF at various temperatures. Figure 48. Cesium-133 chemical shifts as a function of (DBZ7C9)/(Cs+) mole ratio in 30% AN/ 70% DMF at various temperatures. 174 ...20 ...... I .O 2 .0 3.0 4.0 5. O 0.0 [DBZ7C9]/ [Cs+] 175 2| - U! ’ ’_______._. W" (D A 3— E . c1 10 -*'E“D ‘3 ‘5} I) -4>‘4C) ao_3_ c-*BO ‘\ d—>20 e->IO \\ f—>O -91. \ ..IES P. ‘\\\\ g -2 1 1 1 1 1 0.0 1.0 2.0 3.0 4.0 [0327C9]/[Cs+] Figure 49. Cesium-133 chemical shifts vs. (DB27C9)/(Cs+) mole ratio in 70% AN/30% DMF at various temperatures. 176 E "6 r- O. 3 00 ..‘(1 __. 40 °C 30 “l8 "" ? “IO —22 l ' J I I 0.0 .0 .0 .0 .0 [D13 24 C8]/[Cs"’] Figure 50. Cesium-133 chemical shifts vs. (DBZ4C8)/(Cs+) mole ratio in 30% AN/70% DMF at various temperatures. 177 '2l l ‘010 a- g — 0 ->50 °C b->4O c—-30 (ll—>20 e—>IO f->O 0.0 LO 2.0 3.0 [DB 24 cs]/[Cs+] 4.0 5.0 Figure 51. Cesium-133 chemical shifts vs. (DBZ4C8)/(Cs+) mole ratio in 70% AN/30% DMF at various temperatures. 178 Table 52. Formation Constants of (DB27C9.Cs)+ Complex in AN/DMF Binary Solvents at Various Temperatures. Solvent Temperature (°C) Log Kf 10% AN + 90% DMF -10 2.96 i 0.01 0 2.79 i 0.01 10 2.64 i 0.02 20 2.50 i 0.01 30 2.28 i 0.01 40 2.17 i 0.05 30% AN + 70% DMF -20 3.50 i 0.04 -10 3.20 i 0.01 0 3.03 i 0.02 10 2.89 i 0.01 20 2.69 i 0.01 30 2.55 i 0.01 70% AN + 30% DMF 0 3.51 i 0.03 10 3.45 i 0.02 20 3.34 i 0.03 30 3.20 i 0.02 40 3.01 i 0.02 50 2.87 i 0.02 179 Table 53. Formation Constants of (DB24C8.Cs)+ Complex in AN/DMF Binary Solvents at Different Temperatures. Solvent Temperature (°C) Log Kf 30% SN + 70% DMF -10 3.11 i 0.02 0 2.98 i 0.02 10 2.78 i 0.02 20 2.54 i 0.02 30 2.44 i 0.03 40 2.29 i 0.03 70% AN + 30% DMF 0 3.59 i 0.05 10 3.35 i 0.02 20 3.18 i 0.01 30 3.03 i 0'01 40 2.85 i 0.02 50 2.68 i 0.02 180 ’1 a 1. 7 1- x- c .1 5 ~ ‘ 0((nsz7c9.Cs) 111 M1 AN / an: 0111' ‘ 0 (DRZ4C8.C5)‘1n 70: AN / 30: par 5 * (0027C9.Cs)‘1n 30: AM I 70: DH? 0 (0874CR.Cs‘)‘1n 3m 1m / 707 nu: fi (0827C°.Cs)‘1n 10: MI [‘90: nvr 1 1 I 1 l 1 ‘ L —J J 30 3' 32 33 3‘ 35 36 37 as 39 10 g 3 ? $10 . . 4. Figure 52. Van't Hoff plots for complexation of the Cs ion by DB27C9 and DB24C8 in acetonitrile- dimethylformamide binary solutions. 181 .thv mocmummmm 000 00.0 H 00.0 1 00.0 H 00.01 00.0 H.00.01 00.0 H 00.0 00024 0000 00.0 H 00.0 1 00.0 H 00.01 00.0 H 00.01 00.0 H 00.0 020 000 + 20 000 00.0 H 00.0 1 00.0 H 00.01 00.0 H 00.01 00.0 H 00.0 020 000 + 20 000 00.0 H 00.001 00.0 H 00.01 00.0 H 00.01 00.0 H 00.0 020 00:0 "+Amu.0o00000 00.0 H.00.0 1 00.0 H 00.01 00.0 H 00.01 00.0 + 00.0 20 00:0 00.0 H 00.0 1 00.0 H 00.01 00.0 H 00.01 00.0 H 00.0 020 000 + 20 000 00.0 H 00.0 1 00.0 H 00.01 00.0 H 00.01 00.0 H 00.0 020 000 + 20 000 00.0 H 00.0 1 00.0 H 00.01 00.0 H 00.01 00.0 H.00.0 020 000 + 20 000 00.0 H 00.001 00.0 H 00.01 00.0 H 00.01 00.0 H 00.0 020 0000 "+Amo.0000000 000 0002\0000 A00o:\000.00 A00oz\00u.00 00.000 00 0o0 uc0>0om .00 .00 00.000000 .mwnnutz humch 000:9 cam za can MSG CH mmxmamfioo +AmUemUvmmov ocm +Amo.moh~mav Mom mumumemumm 0080c>©050038 .0m manna 182 related to the change in the structure of the solution and the solvent. Hardly any information about the thermo- dynamics of complexation is available in mixed solvents. The only available literature data appear to be those for methanol/water mixtures (47, 122). It was of interest to us, therefore, to perform some thermodynamic measurements in mixed nonaqueous solvents in order to see how the thermodynamics of complexation is affected by the solvent composition. The free energy of the complexation (AG3), is related to the change in the enthalpy (AHE) and the change in the entropy (A83) according to the following expression: AG° = AH° - TAS° . c c c For formation of a stable complex (AGE < 0) there are five possible combinations: (a) AH; < 0 and dominant, TASE > O, (b) AH; < 0 and dominant, TASS < 0, (c) AHé < O, TASé > 0 and dominant, (d) AHS > 0, and TASS > 0 and dominant, (e) AH; < 0, TAsg > 0, and AH; and TASS are the same. Experimental values of AH; and A83 for (DBZ7C9-Cs)+ and (DBZ4C8-Cs)+ systems (Table 54) show that in all cases the complexes are enthalpy stabilized but entropy de- stabilized (i.e. AH; < 0 and dominant, TASE < 0). On the other hand, the thermodynamic parameters, (A63 and A83 values) vary with the composition of solvents. 183 It seems reasonable to assume that the decrease in entrOpy upon complexation is related to a change in the conformational entropy of the ligand. It is known that large macrocyclic ligands such as dibenzo-27-crown—9 and dibenzo-24-crown-8 are rather flexible in the free states, therefore, the negative entrOpy changes may be attributed to the increased ligand rigidity upon coordina— tion. It should be stressed, however, that the conforma- tional change of the ligand is not the only factor govern- ing the change in entrOpy of complexation. Complexation reaction involves not only a change in the solvation of the cation, but also of the ligand. The relative enthalpy and entropy changes can be understood if ligand solvation is taken into consideration. The importance of ligand solvation on the macrocyclic effect has been pointed out by Hinze and Margerum (129) but, at the present time, the information on the interaction of macrocyclic ligands with the solvent molecules is exceedingly sparse. The data collected in Table 54 show that in dimethylformamide-acetonitrile mixtures in the case of (DB24C8-Cs)+ complex, lowering the concentration of dimethylformamide results in decrease in the AG; and AH; vlaues but an increase in the Asa values. In a strong solvating solvent such as dimethylformamide, the solvation of the metal ion (and probably of the ligand) will be stronger than in a solvent of lower solvating ability such 184 as acetonitrile. Therefore, less energy is necessary for desolvation step of the cation (and probably of the ligand) in the case of acetonitrile than dimethylformamide solutions. It would be expected, therefore, that the AH: values for the complexation reaction will be more negative as the acetonitrile composition of AN/DMF mixture in- creases, and the A83 value will increase due to the addi- tional solvent molecules which may release in the reaction. As it is seen from Table 54, while the thermo- dynamic parameters of (DB24C8-Cs)+ are influenced by the solvent and vary regularly with the solvent composition, however, this order is not followed by (DBZ7C9-Cs)+ com- plex, in which the thermodynamic quantities do depend somewhat on the solvent composition but not in a regular manner. This observation may be explained by the con- sideration of the anomaly thermodynamic behaviour of DB3OC10 (70) and DB27C9 (page 126) in complexation with the cesium ion in acetonitrile solutions. It is reasonable to assume that acetonitrile may form a complex in solution with DBZ7C9, such as that reported with 18-crown-6 (107). CHAPTER VI MISCELLANEOUS 185 A. Ionic Solvation of Cs+ Ion in Mixed Solvents a. Results Competitive solvation studies for Cs+ ion were performed in H O-AN, HZO-MeOH, HZO-DMSO binary systems 2 using the cesium-133 NMR signal. For this purpose, the 133Cs chemical shifts were measured as a function of sol- vent composition in these mixtures. The data re pre— sented in Table 55 and the plots of cesium—133 chemical shifts versus solvent composition are shown in Figures 53-55. b. Discussion Study of ion-solvent interaction and preferential solvation of an ion involves the determination of the effect of the solvent on the resonance frequency of the corresponding ion in solution. Some investigations of competitive solvation have been made by several researchers, 59 27 7 205 using Co (130, 131), 23 A1 (132), Li (133), T1 (81, 84, 85, 102), and Na (12, 134-138) NMR techniques. When a solute is dissolved in a binary solvent, the solvation shell of the solute need not maintain the same composition as the bulk solvent, but it may be pre- ferentially solvated by one of component. Frankel et al. (139) have suggested that a convenient measure of the 186 187 00.00 00.0 00.00 00.0 00.00 00.0 00.001 00.0 00.00 00.0 00.00 00.0 00.001 00.0 00.00 00.0 00.00 00.0 00.001 00.0 00.00 00.0 00.00 00.0 00.001 00.0 00.00 00.0. 00.00 00.0 00.001 00.0 00.0 00.0 00.0 00.0 00.01 00.0 00.0 00.0 00.0 00.0 00.01 00.0 00.0 00.0 00.0 00.0 00.01 00.0 00.0 00.0 00.0 00.0 00.0 00.0 00.01 00.0 05000 0020 00 02 05000 0002 00 02 05000 20 00 02 0000 200.00 0000 200.00 020000 200.00 000 1 0020 000 1 000: 000 1 20 00c0>0o0 00x02 00 000000000 0000 500000 00 000000 00002000 0001500000 .00 00000 188 E Q. 3 .15 K) a. O 60 0.0 05 MP. of AN Figure 53. Cesium-133 chemical shifts vs. composition of AN in HZO-AN binary mixtures. |.O 189 0.0 '4.0 '80 'l2.0 "16.0 -200 -24.0 -28.0 “'32.0 8 of '33Cs (ppm) '360 -40.0 l 0.0 0.5 M. F. of Me OH Figure 54. Cesium-133 chemical shifts vs. composition of MeOH in HZO-MeOH binary solutions. LO 190 E O. 3 .15 K) a. C) 60 J 0.0 0.5 MP. of DMSO Figure 55. Cesium-133 chemical shift as a function of DMSO content in HZO-DMSO binary solutions. l.0 191 degree of preferential solvation in a binary solvent, is the bulk solvent composition at which both solvents par- ticipate equally in the contact solvation shell of the solute. It has been suggested that this is the composition in the chemical shift-solvent composition at which the chemical shift of the appropriate nucleus lies midway be- tween the values for the pure solvents (Isosolvation point, I.S.P.). It should be noted that there does not seem to be the firm theoretical basis for this assumption. The isosolvation points for HZO-AN, HZO-MeOH, and HZO-DMSO systems found in this study, are 0.30, 0.48 and 0.55 mole fractions of water respectively. These data suggest that the order of the solvating ability of these solvents towards the Cs+ ion is dimethylsulfoxide > water ~ methanol > acetonitrile. It is interesting to note that, despite the high donor ability of water (D-N = 33, obtained from 23Na NMR measurements, 133), the cesium ion is more strongly solvated in dimethylsulfoxide (D-N = 29.8) than in water. It seems reasonable to assume that the enhance- ment of solvating ability of DMSO in H O-DMSO mixture, may 2 result from breaking up the polymeric structure of this solvent by addition of water (see page 38). The position of the isosolvation points in HZO-MeOH and H O-AN systems 2 indicates that the solvating abilities of water and methanol are approximately equal, but water is a better solvating solvent than acetonitrile. This is not unexpected, 192 Table 56. I.S.P data, Kl/n, and AG"/n for Cs+ ion in the binary mixed solvents Mixture I.S.P Kl/n AG°/n K.J/Mole HZO-AN 0.30 3.09:0.09 -2.83 HZO-MeOH 0.48 1.12:0.01 -o.29 H O-DMSO 0.55 0.48:0.04 1.85 2 193 because the donor ability of acetonitrile (DN = 14.1) is much lower than that of water. Convington et a1 (140-142) have proposed an expres- sion for the equilibrium constant for a solvent-exchange process. The equilibrium constants and the free energy of competitive solvation for the above mentioned systems were calculated using Convington's equation: ---—---0 (l) where 5 = observed chemical shift relative to the resonance of Cs+ in pure solvent A. 6p = total range of the chemical shift between two pure solvents. Kl/n = the geometric equilibrium constant n = the solvation number of Cs+ ion XA and KB = the mole fraction of the two solvents The geometric equilibrium constant, Kl/n, and %— P are obtained from the slope and intercept of the plot of 1/6 versus XB/XA respectively. The free energy of com- petitive solvation AG°/n is calculated by the following expression AG°/n = —RT ln Kl/n . l/n The values of isosolvation point, K (obtained by linear least square, KINFIT program), and AG°/n for each system are given in Table 56. We feel that the high sensitivity 194 133Cs to the solvent en- of the resonance frequency of vironment makes it a useful system for the study of pre- ferential solvation. B. Complexation of the Tl+ Ion by C222 Complexing of the Tl+ ion by cryptand-222 was studied in acetone, acetonitrile, and dimethylsulfoxide 205T1 solutions using thallium-205 NMR technique. The chemical shifts as a function of the C222/Tl+ mole ratios are summarized in Table 57. Since the exchange between the free and complexed thallium ion is slow compared to the NMR time scale, two 205Tl signals (corresponding to the free and the bond Tl+ ion) are observed, as long as the mole ratio of C222/T1+ is between zero and unity (Figure 56). In general, if the exchange rate between the two cationic sites is larger than /7/nAv (Av is the dif- ference between the resonance frequency in Hz of the free and complexed cation), only one population averaged signal isobserved, but when the exchange rate is smaller than /7/nAv, then two NMR signals, corresponding to the free and bonded cation will be observed. Thallium ion forms a very stable complex. The strength of the interaction between the Tl+ ion and C222 205Tl resonance is ob- is shown by the fact that no free served at 1:1 or higher mole ratio of C222/T1+ even in strongly solvating solvent such as dimenthysulfoxide. Table 57. 195 Thallium NMR Study of TlClO with C222 in Dif- ferent Solvents 4 Solvent Conc. Tl+ C222/Tl+ 205T1 Resonance (M) Ratio (ppm) AN 0.005 0.00 -215.69 0.60 —214.23 and 40.54 1.00 40.18 1.88 42.28 AC 0.01 0.01 -183.10 0.50 -l76.73 and 40.96 1.00 38.00 1.65 40.52 4.66 39.82 DMSO 0.02 0.00 319.36 0.48 316.61 and 67.89 0.55 316.28 and 69.75 0.69 314.16 and 69.04 0.84 313.10 and 68.15 0.92 314.20 and 68.52 1.39 67.88 2.21 68.84 3.29 68.69 196 + Free TI 0*- Free Tl Comptexed + JLWL L 1 J 1w lfi 1 1 1 l 1 420 320 220 120 420 320 220 120 20 [c 222]/[T1] = 0.00 [0222] /[T1] = 0.48 L, 1 1 1 1H LA 1 1 1 1 420 320 220 120 20“" 420 320 220 120 20 [C2221/[fi]= 0.69 [C222]/[Tl] . 0.84 LWLvi J l I l 420 320 220 l20 20 [C222]/[T1] = 0.92 Figure 56. Thallium-205 NMR spectra of (C222-Tl)+ complex at various C222/T1+ mole ratios in dimethyl- -su1foxide solutions. 197 Figure 57 provides a graphical representation of the 205 T1 chemical shifts of both free and complexed thallium ion as a function of C222/T1+ mole ratios in AN, AC, and DMSO solutions. It is interesting to note, that the metal ion-complexed resonance frequencies are not identical in these solvents, but the differences are small compared to those for the free solvated ion. This seems reasonable because in (C222-T1)+ where the ionophore (C222) occupies the coordination sites of the thallous ion, the solvent effect would be expected to be much smaller than free thallous ion. + . C. Complexation Studyyof the Na+, and Cs Ions Wlth Cyclo-(tetraethylene-glycol-Z,6-Pyridine decarboxylate) in Various Solvents a. Sodium ion Complexation of the sodium ion with macrocyclic ligand cyclo-(tetraethylene-glycol-2,6-pyridine dicarboxylate) was studied in acetone, acetonitrile, and propylene carbonate solutions. The data obtained from these studies are given in Table 58, and the plots of 23Na chemical shift as a function of the Ligand/Na+ mole ratio are shown in Figure 58. In all three cases the log K value was found to be larger than 5, which indicates f the formation of a very stable complex. As shown in Figure (I) this macrocycle differs from the simple crown ethers because of the presence of 198 35°F "*M-o r1151: 11+ 1:: nuso 1 5° 1- “" 3 c 3 £ fi- cosmmm 11+ 1:: Man E “Munster, C 0. 5‘: ‘ fie 0- m:m.rxrn n 11: AC mm A‘: a. 00 - no - (L—H ram: 11“!!! AC H—H—t FREE 7111:: A11 -...1 o o o l l ‘ 1 k . '-°° 2,00 3.00 ‘ooo 5.00 czzg/fi Figure 57. Thallium-205 chemical shifts of both free and + . . + complexed T1 ion as a function of C222/Tl mole ratio in various solvents. 199 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.001 00.0 00.0 1 00.0 00.0 1 00.0 00.0 1 00.0 00.0 1 00.0 00.0 1 00.0 00.0 1 00.0 020000 +02\0 020000 +0zx0 050000 +0z\0 00 20 00 .0.0_H 00 00 0000>Hom 050000> :0 AmumaaxOQHOOlemchHummlo.mlaoowam1mcmahnummuumpv 000>U mo mocmmmum 0:0 c0 0200.00 0000000wcmsmmuuwa Edeom mo hwsum oHumm 0002 .mm @0009 200 .mu:w>000 050000> :0 00000 000& +02\00000>x0200000 ochH0>m1m.m100000m10c00m500000000IoHomo .0> 00MH£0 HMOHfimno mmIEdeom T0519: 00.0 006 00.0 cod 00.. 00.0 _ _ _ . _ 1 J I. 0(711 JAWP. 11W1 AU 00 1D 1 21011 {1 a 9 .00 000000 (1000):: 201 pyridine ring and the ester groups in its strucutre. Sub- stitution of a nitrogen atom for the oxygen in 18C6 causes a considerable drop in stability constants for alkali metal ions (6). o :N i 0 Ora/To c °) (0 03 w W It has also been noted that addition of two carbonyl groups to 18C6 to form compound (II) causes a considerable dr0p in the stability and cation selectivity (143). Therefore, it is not the presence of the nitrogen or the carbonyl groups which are responsible for increasing the formation constant, but the aromatic ring may change the effect of carbonyl groups on the complexation. b. Cesium ion The data obtained from cesium-133 NMR studies for complexation of the Cs+ ion with the cyclo (tetraethylene- glycol-2,6-pyridine dicarboxylate) in nitromethane, acetonitrile, propylene carbonate, and dimethylformamide solutions are given in Table S9. The plots of 133Cs chemical shifts versus Ligand/Cs+ moel ratios are shown in Figure 59. The values obtained for the stability constants .mmo.o u zommo “0 .ocoo« 202 mm.H m~.¢ em.HH- mm.v oa.m om.v om.mau om.v me.m am.~ mo.o~- mm.m o~.m na.m o~.mau «H.m an.m mv.~ mm.m I mv.m Ha.ma sm.~ 5H.oan ov.~ aa.a sm.a mm.n . sa.a mo.ma mm.H mm.mau am.H vm.s om.H mH.m . m~.H mm.a~ am.H cm.mH- Hm.H ma.o ~H.H am.m . 0H.H mo.q~ om.H va.Hmn ¢~.H >H.m mm.o am.~au Ha.o mo.mm HH.H mm.vmn mo.H Hs.m om.o Hm.¢au mm.o om.w~ Ho.H wa.mmu mo.H mm.¢ mo.o «m.man oo.o mm.am mm.o H~.nm- mm.o ha.v mm.o Ha.o~- em.o H~.m~ em.o m~.m~- mm.o wm.m av.o OH.H~- om.o ao.mm m~.o «H.mm- on.o vm.m ma.o mm.v~- av.o no.0m «m.o oe.mm- Ho.o mh.m mm.o mo.m~- em.o mm.om em.o ma.oe- mv.o vH.~ m~.o mm.wmu mm.o mm.mm am.o av.evu mm.o mm.o ma.o mm.m~- ea.o oa.mm oo.o v~.mm- oo.o no.0- oo.o mm.mm- oo.o Aemcca +mo\a redeem +muxa lemmcw +m0\q Aemmvc +mo\q *za «:2 «man «on mucm>aom moowum> ca coH +mU nufi3 xmamEOU Amumasoconumoac wcflwaummummmuaoomamncmasaummuumuc cacao mo mosum oflumm mac: .mm mHnme 203 40* 20% - 1 1 0-0 1-0 2-0 3-0 4’0 5‘0 [Lig] /[Cs+] Figure 59. Cesium-133 chemical shifts vs. cyclo- (tetraethylene-glycol-2,6—pyridine dicarboxylate)/ + . . . Cs mole ratio in various solvents. 204 of the complex in NM and AN solutions (assuming that only 1:1 complex is formed) are log Kf = 3.29 i 0.19 and 1.93 i 0.10 respectively. These results are not un- expected, since acetonitrile (DN = 14.1) has a stronger solvating ability than nitromethane (DM = 2.7). It is interesting to note that in dimethylformamide and propylene carbonate solutions, the addition of the ligand to Cs+ ion solutions results in a downfield shift as the Ligand/Cs+ mole ratio increases from 0 to 1.5, then further addition of the ligand gives an upfield shift. The slight upward displacement of the 133Cs chemical shift beyond the mole ratio of 1.5 is an evidence for a two-step complexation reaction, however, the small change in the 133 Cs chemical shifts in these solutions makes quantita- tive analysis of the data impossible. D. Complexation of the Tl+ Ion by BlSCS Complexation of the T1+ ion by monobenzo—lS- Crown-S was studied in acetone, dimethylsulfoxide, pyridine, and dimethylformamide solutions. The resulting data are given in Table 60 and the mole ratio plots are shown in Figure 60. In dimethylformamide and pyridine solutions the plots show little curvature without any break at any mole ratio, which is an evidence for the formation of a very weak complex between Tl+ ion and BlSCS in these sol- vents. Computer analysis of the data for the (BlSCS-Tl)+ system in DMF and FY solutions gave the log of formation 205 .mfio.o n acaoae mo .ocoon .mmoo.o n «03035 00 .ocoom ow.3vm no.0 mb.mom mm.m m~.¢ n mm.m oo.mm~ mm.m 3m.m3 mv.v m3.w3~ mm.m mm.mvm mn.m me.momu ms.v o>.mm os.m oa.vm3 n~.m ea.ovm oa.m ©3.ma3u w~.v em.vv oo.m 3m.oma 3m.~ mm.mem mm.m mm.mnau mm.m mm.¢m mm.~ mm.m33 mm.3 ~m.ovm mo.m vv.om3u No.~ mm.nm oa.m 3m.mm vm.3 av.nmm om.3 m3.~v3u m3.m 3m.~m am.3 mm.mm o¢.3 mo.mmm mm.3 mo.mm3n ma.3 mo.mm mm.3 ve.mw m3.3 v3.mmm «3.3 ov.m~3- mm.3 m3.mm a3.3 ~«.vm mm.o mo.¢mm mm.o 3o.-3: om.3 om.ooa oa.o mm.~m hm.o mm.mmm mh.o 3o.a~3: am.o mm.~33 mm.o om.m3 nm.o n~.3mm e¢.o ov.mv3u om.o om.m~3 $3.0 ma.» 3~.o m3.omm s3.o m3.mmau mm.o m3.mm3 oo.o mm.m n oo.o mm.mmm oo.o mm.om3n oo.o Edd: +2.3 Edd; +2.3 Edema +2.3 Ema: +3.3 mED Mm n mucw>aom moofium> CH mUmHm an coH +HB m0 coflumxwamfioo MOM mama oflumm mHOZIHMflnm HMOflEmno manna 206 350 {’ omso W Pv 1 no- 230- _ 4wL A E In- D. 3 -“0r— w 1 1m- 4w 50*” ' AC 1 1 ' 0 '0 230 310 £ 0 4 .10 u- 1 J. 1 , J DMF. om vc 2m to 4m ‘ so 5% 7% $0 15C5/1'. . . . . + Figure 60. Thallium-205 chemical shifts vs. (15C5)/(Tl) mole ratio in various solvents. 207 constant 1.01 i 0.01 and 0.92 i 0.03 respectively. These solvents (nitrogen donor, soft base) have a strong sol- vating ability towards Tl+ ion (soft acid) and can compete quite successfully with the ligand in the complexation process. As indicated in Figure 60, no significant variation in the 205Tl chemical shift is observed upon addition of the ligand to the thallium salt in dimethylsulfoxide solutions. In this case the solvation 0f the Tl+ ion must be strong enough to prevent complexation reaction. The addition of BlSCS to a solution of T1+ ion in acetone (a solvent with a medium donicity; DN = 17.0) re- sults in a downfield shift followed by a break at the 1:1 Ligand/Cation mole ratio; beyond the mole ratio of 1, an upfield shift is observed astflmaconcentration of the ligand is increased. This behaviour shows that at least two com- plexes with the respective stiochiometry of 1:1 and 2:1 (sandwich) are formed in acetone solutions. It is obvious that the thallium ion is too large (ionic diameter = 2.8 A°) to fit into the ligand cavity, but it may lie above the ligand ring. Support for this observation is the stepwise formation of (B15C5-Tl)+ and (B15C5)2-Tl+ complexes in this solvent. As we expect, comparison of our results obtained 205 from T1 NMR studies on complexation of the Tl+ ion with 18C6 (page 74) and BlSCS in nonaqueous solvents demonstrates 208 the weaker complexing ability of B15C5 than 18C6 for the Tl+ ion. A possible reason for the lower stability of the complex formed between BlSCS and the Tl+ ion is that this cation is too large to match the ligand cavity since the cavity diameter of the macrocycle (~ 2A°) is smaller than the cation diameter (2.8 A°), but the cavity of 18C6 (cavity diameter ~ 3A°) accommodates well Tl+ ion. An- other factor which may influence the decreasing of the complexation strength of BlSCS compared to 18C6 is the number of donor atoms which in the case of BlSCS is less than 18C6. E. Thermodynamic Study of Complexation of DB21C7 with the Cs+ Ion in DMF and PC In order to determine the thermodynamic quantities for the complexation of the Cs+ ion by DBZlC7 in dimethyl- formamide and propylene carbonate solutions, the variations of the cesium-133 chemical shifts were determined as a function of temperature. All the 133Cs chemical shifts measured at different temperatures are listed in Tables 61 and 62. The plots of 133Cs chemical shifts as a function of DBZlC7/Cs+ mole ratio are shown in Figures 61 and 62. The formation constants of the (DB21C7-Cs)+ complex at various temperatures are given in Table 63. As shown in this table, the stability of the complex varies significantly with the temperature, suggesting that the equilibrium be- tween the Cs+ ion and the ligand is temperature dependent. 209 Nw.vl Hm.ml mm.al wN.OI hm.o mm.m mb.ml vw.ml wh.HI vm.ou mh.o mN.m NN.mI mm.MI om.HI mV.OI mh.o mm.m oh.ml ma.vl MH.NI Nb.OI mm.o on.a mm.ml mN.vI ON.NI om.ot m¢.o H¢.H mm.ml oo.vu hw.ml HH.HI mN.o mo.H mv.wl mo.ml 0H.ml Nv.HI Ho.OI mh.o wH.hl mm.ml mm.MI mn.al . wm.ol mv.o oq.hl mm.ml oo.ml vo.NI mv.0I NN.o hw.hl mm.mn mm.ml mN.NI mm.OI oo.o Op 00 om ov om +mU\.H Uo wusumquEmB .mmuoumuwmema ucmummwflo um sz ca zummu Emoo.o :ufi3 xmameoo hoammo m0 zpsum ofiumm 0H0: .Hm magma 210 hm.val nm.man mm.mal wo.aal vm.oau mh.m mv.mHI NH.vH| nm.mH| hm.aal hH.HHI vo.m om.mal mw.¢an Ha.mal mo.mH| mo.HHn mm.m hm.wal om.ma| mm.mal wH.mHI mo.HHn mm.H mm.>a| H~.mal mm.val om.ma| vw.aan Hm.H om.om| mm.m~| mm.man HH.vHI ma.ma: 0H.H mm.mml mm.om| mh.mau m>.wH| Hm.man vm.o no.mml ¢>.mm| wo.v~n vm.mml H¢.Hml mm.o vo.am: mh.mm| Hm.mml mm.m~: mm.mm| >m.o mm.vml mv.mm| HH.mmu ah.om| on.mmn mH.o mm.mmn No.5ml mm.mm: vh.vmu Hm.mm| oo.o om llom ow om om Uo ousumummfims +mU\A .mmuoumummfime ucmumMMHo um um a3 zommo :moo.o nuflz xmaceoo scammo mo mwsum oflumm mac: magma 211 30 °C C1£>F" ‘4C) 50 -20/ 60 E O. 3 -4.0 G!) '7C) -€;(DL/t////Jy,AV’I.I’dr”””””"”v—"—." '1.C) (1!) LC) 25() El!) [03 2| C7]/[Cs+] Figure 61. Cesium-133 chemical shifts vs. (DBZlC7)/(Cs+) mole ratio in DMF at various temperatures. ‘11) 51() 212 '8 . . 20 ‘C - 3O '2 F— 40 5O 60 ..EBC)._- E 8, -24 — 0° _ ll/ / '32 //l ' / ’l “'36 ’1 ‘40 I J I J 0.0 LO 2.0 3.0 4.0 [082:07]/[Cs+] Figure 62. Cesium-133 chemical shift vs. (DBZlC7)/(Cs+) mole ratio in PC at various temperatures. 213 Table 63. Stability Constants of (DBZlC7 Cs)+ Complex in DMF and PC at Various Temperatures. Solvent Temperature (°C) Log Kf Dimethylformamide 70 1.93 i 0.12 60 2.06 :_0.06 50 2.43 i 0.09 40 2.51 i 0.05 30 2.78 i 0.05 Propylene Carbonate 60 3.16 i 0.02 50 3.25 i 0.05 40 3.47 i 0.03 30 3.80 i 0.05 20 4.12 0.13 I+ 10!- s— " 0 <3 7» x c .l s- 5- (JPc tr 00m: 4 i 4 v 4 L g 28 29 so 31 32 33 34 as 1 3 -T-%IO Figure 63. Van't Hoff plots of In Kf vs. i for T (DB21C7-Cs)+ system in PC and DMF. 215 mw.m + mm.mHI vm.~ + wH.HNI vH.H + «H.HHI No.0 + wm.oal no.0 + hm.ml no.o + mm.ml mo.o mo.o + mh.m 000000000 mcmammoum mpflfi0fiuomawnumeflo Ace 0302\Hmov om< Amao:\300 xv 034 Amaoz\amo xv 00< Aooomv x meg ucm>aom um 000 min cw xwamfiou +AmU.>UHNmQV MOM mumumfi0u0m 0flE0c>UOE0009 .vw 0HQ0B 216 Van't Hoff plots of ln K versus l/T are shown in f Figure 63. The enthalpies and entropies of the complexa- tion were obtained by the usual manner from the slopes and intercepts of these plots. Table 64 summarizes the thermo- dynamic quantities of the (DBZlC7-Cs)+ complex in propylene carbonate and dimethylformamide solutions. Investigation of these data indicates that in both cases the complex is enthalpy stabilized but entropy destabilized. It seems reasonable to assume that the negative entropy upon complexa- tion arises from a change in the conformational entropy of the ligand. Since dibenzo-Zl-Crown-7 is a flexible ligand (because of its large size), it will lose a large degree of freedom in the complexation process. APPENDIX APPENDIX DETERMINATION OF COMPLEX FORMATION CONSTANTS BY THE NMR TECHNIQUE; DESCRIPTION OF COMPUTER PROGRAM KINFIT AND SUBROUTINE EQUATION 217 The equilibrium for a 1:1 (metal ion:ligand) com- plexation reaction can be written as: M+L:ML (l) and the concentration equilibrium constant is given by the following expression: CM ° CL where C refers to the molar equilibrium concentrations. Since in this case, the complexation reaction does not re- sult in a separation or combination of charges and the concentrations of the reactants are very low (~ 0.01), we only consider the concentration formation constant. Under these conditions, the values of concentration con- stant will closely approximate the thermodynamic value. Assuming that only cation-ligand interaction are important and the rate of exchange of the metal ion be- tween the two sites (free cation in the bulk solution and complexed cation) is fast on the NMR time scale, the ob- served chemical shift of M (éobs) is a weighted average of the characteristic chemical shift of free cation and complexed cation: 5obs = 6MXM + GMLXML (3) 218 219 where 6M and 6ML are the chemical shift of the free an complexed M respectively, XM and XML are the fractions of the cation in the free and complex states. Therefore 6 = obs 6MXM + 6ML(1 ‘ XM) = XM(5M ‘ GML) + 6ML (4) The analytical concentration of M is: T _ CM — CM + CML (5) and the analytical concentration of ligand is: T - CL ' CML + CL (6) Therefore _ T _ CML - CM CM (7) and _T_ T_ CL — CL (CM CM) (8) Then T (CM - CM) K T T (9) (CM)(CL - CM + CM) CM is solved in (9) T T _ T _ CM(CL - CM + CM)K — CM CM 2 T T T_ KCM + (KCL KCM + l)CM - CM - 0 c = 5— {-(KcT — KCT + 1) + M 2K L M — T \jQKCL - Kc; + 1)2 + 4KC§)} (10) 220 Since CM cannot be negative, therefore the positive root is chosen: 2 2 (KcT-KcT-1)+\/chT +chT -2K2CTCT+2KCT+2KCT+1 _ M L L M L M L M By substitution of C from (11) in equation (4) we can M derive the following expression: 2 2 _ T _ T _ 2 T 2 T _ 2 T _ T éobs — [(KcM KCL 1) + (K CL + K cM 2K CM cL T T L 6M ' GML + 2KC + 2KC + 1)2](————————) + 6 (12) L N T ML In order to fit this equation, two constants and two parameters are used in the FORTRAN CODE: II o) ZFBE U(l) U(2) K 6 CONST(1) M ll 0 CONST(2) The input variables are the analytical concentration of the ligand (XX(l)) and the observed chemical shift (XX(2)). The data input includes the control cards and the NMR data. The first card contains the number of data points (columns 1-5 (F.15)), the maximum number of iterations to be performed (columns 10-15 (F.15)), the number of constants (columns 36-40 (F.15)), and the maximum value of (A parameter/parameter) for convergence to be assumed (0.0001 works well) in columns 41-50 (F10.6). The sub- sequent cards include: (l) (2) (3) (4) 221 a title card a card containing the values of CONST(1) columns 1-10 (F10.6) and CONST(2) columns ll-ZO (FlO.6) a card containing the initial estimates of the un- knowns U(l) and U(2), in columns 1-10 and 11-20 (F10.6), respectively The fifth through N cards contain XX(l) in columns 1-10 (FlO.6) variances on XX(l) in columns 11-20, XX(Z) in columns 31-40 (FlO.6) followed by the same parameters for the next data point. The subroutine EQN for calculations of K (forma- tion constant) and 6ML (chemical shift of the complex) is listed below: SUBROUTINE EQN not.1v!.5AD.xxvco.~noi.u?vno.~nuux.x.u.11:: A9.£Psol V°.xx.nx7v9.0II .roo.to.ru.o.zL.to J.v.Ov.v£C!.Ncst.co~sv.~otr.JoAr.~097.Ln°!. L _—~< o J. UN- 19 26 H H H UMfl -¢33£-(D~CK3<3‘KJ DI 3-ll!<