'V— rw a- “ v.— ' -"‘7 Y y-v—A— _‘ VA ”g vaw~ - —_v‘ — . 5 .17: A ii; ti" 1;: '5‘ '. 7 fl?" 1‘ I I , " 'V ‘ I ' 'v . "N. ‘.'.".f ' . YIN: 4 y :51. «IA ‘1 . . l I ' '2 -"$.' ‘.-‘ .'. " .I .- ‘ I. '2. " + w. I "';ia:=~. I - I" ’0‘ “‘ I": I . . ‘r V . u.? c ‘. In: I.'. { 4! ., C, v. “ I \ "it . '- ‘ . "(3‘ NM'; ' ' ' I. "5: g" I O 5‘ . I Ex}. I "' "V‘: 3:25: il-l‘rg.’{‘ :‘fl'fih‘: fin.“ ‘24-. LAK‘. .::.' 33% , L ”1...“. ‘94,}:“hu’ fl . 5‘. c ‘5‘ .~.- - . I‘v‘) ff: 2' ‘|\'r?nu{"\‘.'1: h) ' ‘1 ff. I 1‘. 4.: :u - £325.?! (<57 5 . . _, “a?! qos‘ncoq? (x: [I I . ,. :‘VI \ R'R'Lnjnk‘wf‘lrfl'hu "II. ‘ J - 4:“ ,4? '5‘ (If. .H .- .I.“-_‘". '.‘.. .1" “f "'5“ iz‘iJI-1¢"r§23f~§-'9I“’§ ‘ “(4'11 "C'c‘yfw "27:1 -.'.“I‘n'- ,. 9 . ._. _ mun-Julmfl’. fl'vi“ hwy“? I ,. u"! .- 1'. ‘v ‘h -,...:I..I.I~« wan = “- . ei..~-A.;.;~=Iu '13-- 'Pt‘ j; #;"~"Z‘§’7§ai%’ ' 9;: .- -9 , . ~f \«1 ”Q33?“- "' A“ at?“ \ ‘- ‘é', . .r L\ _ l kglt‘fim‘1‘35w5-fib‘ ‘59.! I? us- if“ I IMM ia‘fiafififiagéfi "me-J ‘ ' \ .l w I ~ ‘ I ' \ III ; :' 'r‘ufi' >a‘ 41‘“... Q . .. -. . .-.-. ‘1' 5‘": 3%" b‘M'-'-}a:- ‘g‘Q'L “'43::‘53'15 ‘}?%3(€)D~Qflfllav§i€a~ .Jéw’, imr‘ifi, ”v.33; ., '.\T‘Y."'"§ _ “v: ":?¢;‘ifiiu’»‘§€u 53533;"?- . 1'." 'f‘ " " “M. '." r. .I. an . .. A ~ 3‘ I It: ,4 ‘"13m"?7I~¥‘II‘3~“'L'v"‘°.p" {131’ ‘39; LFWzW ‘ ’x‘ 1“. I}. I‘ 3.... ‘I'TU-‘f" 1 ._ :‘ . . v 1.2. W“? Fri. 13:3"$%<~:‘~’I'.I 3.; > "' 5‘. -' ‘ Intru- V . I: ‘y fir.‘\ r? “w . ' a - z“ .I ”aw-w. 1,1,},- fi; "*2; 3%” A In» w «.3 " 1 ~* :m': 9-5.... .- ‘ .‘r p A b1 " I" a . ‘ m- 3.311%"; _| _ . ';¢.+fg,~.-:tfir” ' "1’:— “":\"‘PI '1 ‘6 , __ U n.1f-’ I 1 >‘ ‘ ”U viifgn‘: ’ nH-‘w £53?” Jdv'ué'.‘ .. ;..'-.:: {-2491 J. ., Is" __ ‘3’. ‘3 {xi {3' "33.53;” g ‘u'. "‘ $51.43, -~ - I W W ”P1: 0‘ I Y” t TC? J",- I N IAO£ I '.; . .. u— . 395%” ,-:-.r 7} l \ . .‘f .1.“ J - of.‘ 3:41;,“ 1‘“) J J‘, if 4. 5‘” ‘3 l L». “ L .1 . .. ;. ‘3“ 1', " 33:}...‘4m ', “my . -._ I WM“? - -.‘-? .‘ J V214! t. 113'”: "MG '1" g , "1"”, \ ‘fi > I f - u, . .I L , 1 . I...” . .. , v I ~ I I . _ 7- -. ‘ ~ -..:w.u.v. , I l ‘ ' I'. v .‘ . ‘ . H , , .y ' ., t __ . . . . . ‘ ,. ,7” >7 -. :1 .‘l. W?“ -. -- r - > II.’“.. u .2 : . . m. , ~ ‘ ' 7 "3- ‘ - I I".' ‘r'. M 4 I .I 'I‘, " " "' " - 3:1.“ ' 2g " Igpv‘q. . ' '(I‘ l .1? V ‘ I in”: 1' ' _ n... . _. .. . ', ' o '.'\ "‘I . .i 1‘ .‘y.."OJ '1“ 0.. N ‘I‘ .qqu 9 u V-Vd‘ , .. . v. _. .fir, .V'.’ . ‘ ‘1 l I fly I‘ll' I“. r‘ ua.l..'. W I. .I I ... 1‘ yuan” an. N 3., t‘ZI :. . . ' ' 'I JIV'fl'c: 2' I '1. .1 ' I .1415 lll'vr“, . '1‘.’ . . '0. I'r’“ 45"“ “"1? -- ‘A ‘fil‘v . . ,‘ "I.I ,. .. ,A _ t ,.o ,v‘vl I ‘ _‘I’ 1'"... . w ,7- I V . ‘3, ““‘I"' I". "".';II 't'; 'r'” 1J.:'.I‘ I.. .v‘ .' . .fl. '5 ' 1 N‘I.‘ ‘."' ‘. . ~ ",.'IIII.I'I, '..‘.I|). v I»... L: ::,.-..-.'IJ'.. "'31". ’I I. .'r .._'I . . - . I '1':"l II"". Ira. rr' 'er .‘. ‘ .r'l ‘1'» I\ ‘ ‘ A .1~.‘.“"‘ " "In .' .n'l‘“ I,. ‘. ..jl '.-'- / I 3 h. l j." ' ."I g5“ " 3.3%." .5. '. k," ' ’ ' ‘ swéfi‘éflk " I " . I ’ III»; ..—.«21«:I~.-...a~ v.5 mmefi-‘I‘tw _ _ I Ami: %B'&‘7"$§IIIII&?§$§£S&21’ . I222. , . I 4 l I‘ I ‘\I .‘ u” II I ~". "'u A ' W" i Juli: "2“?“ J I "‘l‘ "I I. .‘l‘ )‘f"‘”,"'.|""$ . 1". 1.1%{13'4‘1 s'é'a’fllflw-th‘ >"‘I~=‘Inl.;Ir'-t‘.: 1111 111 11an lllll ll: at THESIS 3 301096 7176 ‘ LEBRARY Michigan 5mm 1 University fiv— W This is to certify that the dissertation entitled J 3‘ ‘ /Z"/e¢ [ten/$073 0 I C futon {*4/97 (0141/ 444793,, (76‘ c172)»; é/Olfz‘ Na‘f Cnc/ (25‘ 77 12.45 presented by 2nd e C) men 8 +r c S 5 £77 has been accepted towards fulfillment of the requirements for EA ob. degree in (7 [181/74/5 +7 //\ 599/1er 1/237 Major professor Date //-/9’ 99/ MSU is an Affirmative Action/Equal Opportunity Instilun'on 0- 12771 v- my] .Péfll’fl'wlg-Jfiflfllfll-fi= i:3\. Place in book drop to LIBRAHJES remove this checkout from Jmnaucufimmc your record. FLNES will be charged if 366% is returned after the date Stamped below. 1 mi” (‘1' m I. l; ‘ 1 } MECHANISMS OF CROWN ETHER COMPLEXATION REACTIONS WITH Na+ and Cs+ IONS By Bruce Owen Strasser A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1984 ABSTRACT MECHANISMS OF CROWN ETHER COMPLEXATION REACTIONS WITH Na+ and Cs+ IONS BY Bruce Owen Strasser The kinetics of complexation of sodium ion with the ligand lB-crown-6 (18C6) were studied in six solvents or solvent mix- tures by a complete sodium-23 nuclear magnetic resonance line- shape analysis. The salt used was sodium tetraphenylborate or the thiocyanate. The solvent systems were tetrahydrofuran, methanol, propylene carbonate, methanol-tetrahydrofuran mix- ture (40-60 mole %) and propylene carbonate-tetrahydrofuran mixtures (20-80 and 60-40 mole %). In tetrahydrofuran solutions, the exchange mechanism is determined by the degree of contact ion pairing of the sodium salt and of its complex with 18C6. When SCN- is the counterion the solvated and complexed sodium ions form contact ion pairs and higher ionic aggregates. This ionic association reduces the charge—charge repulsion of the sodium ions in the transi- tion state of the bimolecular exchange mechanism and allows this process to predominate. When BPhZ or AlEtZ are the counterions, contact ion pairing is minimal and the predomin- ant exchange mechanism is the dissociative process. Bruce Owen Strasser In methanol, in the methanol-tetrahydrofuran mixture, and in the 20-80 mole % propylene carbonate-tetrahydrofuran mix- ture, the predominant exchange mechanism is the dissociative process. However, in propylene carbonate and in the 60-40 mole % propylene carbonate-tetrahydrofuran mixture, the pre- dominant exchange mechanism is the bimolecular exchange process. A correlation has been found between the Gutmann donor number of the solvent and the free energy of activation for the dissociative step in systems in which the dissociative mechanism has been found to predominate. In those cases in which the bimolecular mechanism is predominant, the free energy of activation for the process is independent of solvent. The kinetics of complexation of cesium thiocyanate with the ligands dibenzo-Zl-crown-7 (DBZlC?) and dibenzo-24-crown- 8 (DB24C8) were studied in acetone and in methanol solutions by a complete cesium-133 nuclear magnetic resonance lineshape analysis. In all systems, the bimolecular exchange process has been found to be predominant. Isosolvation studies of sodium ion in methanol-tetrahydro- furan and in propylene carbonate-tetrahydrofuran solutions have been done using sodium-23 nuclear magnetic resonance chemical shifts. The salts were sodium tetraphenylborate and sodium perchlorate. In all systems, the solvent with the higher Gutmann donor number has the lowest concentration at the isosolvation point. This Thesis is Dedicated To My Parents ii ACKNOWLEDGMENT Sincere gratitude is extended to Tom V. Atkinson whose computer wizardry has greatly reduced the time and frustra- tion needed for the completion of this dissertation. Your patience, dedication, and friendship have not gone unap- preciated. Many thanks go to the Popov group. Our long philosophi- cal discussions, whether over coffee or wine, have made my stay here quite memoriable. One cannot forget the dedication of the "happy hour gang". The weeks flew by in anticipation of Fridays at 5:00 p.m. Where do you want to go? The efforts of the nmr group - Klaas, Kermit, and Long - must be fully acknowledged. Your contributions to the nmr facility and the work presented in this thesis have not gone unnoticed. Finally, thanks Alex. Your guidance and friendship throughout my stay here at Michigan State are warmly appreciated. Don't forget to save room for (Meg's) dessert. iii Chapter TABLE OF CONTENTS LIST OF TABLES . O O O O O O O O O O C 0 LIST OF FIGURES O C O O O O O O O C O O 0 LIST OF ABBREVIATIONS. . . . . . . . . . CHAPTER 1. HISTORICAL REVIEW. . . . . . Introduction . . . . . . . . . . . . Complexation Kinetics of Alkali Cation- Crown Ether Complexes. . . . . . . . CHAPTER 2. EXPERIMENTAL PART. . . . . . A. Salt and Ligand Purification . . B. Instrumental Measurements and Data Handling. . . . . . . . . . l. 2. 3. 6. CHAPTER 3. Instruments. . . . . . . . . References and Corrections . Temperature Calibration and Centre]: 0 O O O O O O O O 0 O Linewidth Measurements . . . Data Acquisition and Signal Processing . . . . . . . . . Data Transfer. . . . . . . . MISCELLANEOUS. . . . . . . . A. Ion-Pairing in Tetrahydrofuran Solutions. . . . . . . . . . . . Introduction . . . . . . . . . . Results and Discussion . . . . . l. Conductance Measurements 2. Infrared Studies . . . . 3. NaAlEt4-18C6 Exchange in iv Page vii xii xix 21 22 24 24 24 '27 3o 32 33 3s 36 36 38 38 40 42 Chapter Conclusions. . . . . . . . . . . . . . . B. Isosolvation Studies . . . . . . . . . . Introduction . . . . . . . . . . . . . . Results and Discussion . . . . . . . . . CHAPTER 4. KINETICS OF COMPLEXATION OF SODIUM ION WITH 18-Crown-6 IN NONAQUEOUS SOLVENTS O O O O O O O O O O O O O O A. Introduction . . . . . . . . . . . . . . B. Choice of Solvents and Salts . . . . . . C. Results and Discussion . . . . . . . . . 1. 9. CHAPTER 5. Measurements in the Absence of Exchange. . . . . . . . . . . . . Fourier Transform NMR Exchange Equations. . . . . . . . . . . . . . Mechanisms of Exchange . . . . . . . Results in THF Solutions . . . . . . Results in Methanol Solutions. . . . Results in PC Solutions. . . . . . . Comparison of Results in Neat Solvents . . . . . . . . . . . . . . Results in MeOH-THF Mixture. . . . . Results in PC-THF Mixtures . . . . . COMPLEXATION KINETICS OF CESIUM ION WITH DB 21C? AND WITH DBZ4C8 . . . . . IntrOduction O O I O O O O O O O O O I O O 0 Results and Discussion . . . . . . . . . . . A. B. Measurements in the Absence of Exchange . . . . . . . . . . . . . . Kinetic Results. . . . . . . . . Page 42 45 45 47 54 55 55 59 59 77 84 86 98 105 107 116 121 130 131 132 132 142 Chapter APPENDICES APPENDIX A - DATA TRANSFER PROGRAMS FOR THE WH-180 NMR APPENDIX B - FT-NMR TWO-SITE EXCHANGE EQUATIONS MODIFIED TO INCLUDE LINE BROADENING AND DELAY TIM O O O O O C O 0 APPENDIX C - SUGGESTIONS FOR FUTURE WORK . REFERENCES vi Page 155 159 165 169 Table LIST OF TABLES Page Exchange Mechanisms Postulated or Observed for Alkali Metal Ion-Crown Ether Complexation . . . . . . . . . . . . . 5 Kinetic Parameters for Relaxation of 18C6 in Various Solvents . . . . . . . . . . 8 Kinetic Parameters for Com- plexation of Na+ Ion with Crown Ethers at 25°C . . . . . . . . . . . . . . . 13 Kinetic Parameters for K+ Ion Complexa- tion with 18C6 in Various Solvents at 25°C . . . . . . . . . . . . . . . . . . . . 15 Kinetic Parameters for Complexation Kinetics of Cs+ Ion at 25°C. . . . . . . . . l7 Kinetic Parameters for Complexation of + Ion with Several Crown Ethers in Cs Methanol and in Acetone Solutions at 25°C . . . . . . . . . . . . . . . . . . . . 18 Temperature Calibration of the WH- 180 Spectrometer . . . . . . . . . . . . . . 28 Temperature Calibration of the DA-60 Spectrometer . . . . . . . . . . . . . . . . 29 vii Table Page 9 Equivalent Conductances of Various Tetrahydrofuran Solutions at 25°C. . . . . . 39 10 Sodium-23 Chemical Shift vs. Composition of THF/PC Binary Mixtures at 25°C . . . . . . . . . . . . . . 48 ll Sodium-23 Chemical Shifts vs Com- position of THF/MeOH Binary Mixtures at 25°C. . . . . . . . . . . . . . . . . . . 49 12 Selected Properties of Some Solvents at 25°C. . . . . . . . . . . . . . . . . . . 57 13 Sodium-23 Chemical Shifts and Relaxa- tion Rates of NaBPh4 and of its Com- plex with 18C6 in THF Solutions. . . . . . . 60 14 Sodium-23 Chemical Shifts and Relaxation Rates of NaSCN and of its Complex with 18C6 in THF Solu- tions. . . . . . . . . . . . . . . . . . . . 61 15 Sodium-23 Chemical Shifts and Relaxa- tion Rates of NaBPh4 and of its Com- plex with 18C6 in PC Solutions . . . . . . . 62 16 Sodium-23 Chemical Shifts and Relaxa- tion Rates of NaSCN and of its Com- plex with 18C6 in MeOH . . . . . . . . . . . 63 17 Sodium-23 Chemical Shifts and Relaxa- tion Rates of NaBPh4 and of its Com- plex with 18C6 in a 40-60 mole % viii Table Page MeOH-THF Mixture . . . . . . . . . . . . . . 64 18 Sodium-23 Chemical Shifts and Relaxa- tion Rates of NaBPh4 and of its Com- plex with 18C6 in a 80-20 mole % THF-PC Mixture . . . . . . . . . . . . . . . 65 19 Sodium-23 Chemical Shifts and Relaxa- tion Rates of NaBPh4 and of its Com- plex with 18C6 in a 60-40 mole % PC-THF Mixture . . . . . . . . . . . . . . . 66 20 Sodium-23 Results for Solvated Na+ in Selected Solvents . . . . . . . . . . . . 73 21 Sodium-23 Results for Complexed Na+ in Selected Solvents . . . . . . . . . . 74 22 Mean Lifetimes as a Function of Temperature for the System NaBPh4- 18C6 in THF Solutions. . . . . . . . . . . . 87 23 Mean Lifetimes as a Function of Temperature for the System NaSCN- 18C6 in THF Solutions. . . . . . . . . . . . 88 24 Kinetic Parameters for the Com- plexation of Na+ with 18C6 at 25°C . . . . . 92 25 Weighted Average List of Kinetic Parameters for Complexation of Na+ with 18C6 at 25°C. . . . . . . . . . . . . . 95 ix Table Page 26 Kinetic Results for the System NaAlEt4-18C6 in THF Solutions at 25°C . . . . . . . . . . . . . . . . . . . . 97 27 Mean Lifetimes as a Function of Temperature for the System NaSCN- 18C6 in MeOH Solutions . . . . . . . . . . . 100 28 Kinetic Parameters for the Complexa- tion of NaSCN by Several Crown Ethers in MeOH Solutions at 25°C . . . . . . 103 29 Mean Lifetimes as a Function of Temperature for the System NaBPh4- 18C6 in PC Solutions . . . . . . . . . . . . 106 30 Mean Lifetimes as a Function of Temperature for the System NaBPh4° 18C6 in a 60-40 Mole % THF-MeOH Mixture. . . . . . . . . . . . . . . . . . . 117 31 Mean Lifetimes as a Function of Tem- perature for the System NaBPh4-18C6 in a 20-80 Mole % PC-THF Mixture . . . . . . 122 32 Mean Lifetimes as a Function of Tem- perature for the System NaBPh4°18C6 in a 60-40 Mole % PC-THF Mixture. . . . . . . . 123 33 Cesium-133 Chemical Shifts and Relaxa- tion Rates of CsSCN and of its Complexes with DB21C7 and with DB24C8 in MeOH Solutions. . . . . . . . . . . . . . . . . . 133 Table 34 35 36 37 Cesium-133 Chemical Shifts and Relaxa- tion Rates of CsSCN and of its Complex- ing with DB21C7 and with DB24C8 in AC Solutions. . . . . . . . . . . . . . . Mean Lifetimes as a Function of Tem- perature for CsSCN-Crown in MeOH Solu- tions. . . . . . . . . . . . . . . . . Mean Lifetimes as a Function of Tem- perature for CsSCN°Crown in AC Solu- tions. . . . . . . . . . . . . . . . . Kinetic Parameters for the Complexa- tion of Cs+ with Several Crown Ethers in AC and in MeOH Solutions at 220°K . xi Page 135 143 144 151 Figure LIST OF FIGURES Page Some synthetic and naturally occurring macrocycles . . . . . . . . . . . . . . . 4 Types of ion pairs in solution. . . . . . 37 Room temperature infrared spectra of tetrahydrofuran solutions containing NaSCN. I) 0.126 M NaSCN + 0.140 M 18C6: Spectrum of Na+-1BC6-Ncs'. II) 0.074 g NaSCN, (a) Na+-Ncs'-Na+; (b) Na+-NCS-; (c) (Na+-NCS-)2. Assign- ments from Reference 56 . . . . . . . . . 41 Sodium-23 nmr of a solution containing [NaAlEt4J/[18C6] z1.9 in THF at several temperatures. . . . . . . . . . . . . . . 43 Sodium-23 chemical shift XE' composition of THF/PC mixtures. (o) 0.1 M NaBPh4; (x) 0.1 M NaClO4; (*)(+) isosolvation points. 50 Sodium-23 chemical shift XE‘ composition of THF/MeOH mixtures. (o) 0.1 M NaClO4; (x) 0.1 M NaBPh4; (*) isosolvation point. 51 Semilog plots of l/T2.X§° reciprocal temperatures for tetrahydrofuran xii Figure 10 11 Page solutions containing solvated and 18C6 complexed Na+ ion. . . . . . . . . . 68 Semilog plots of l/T2 XE' reciprocal temperatures for solutions containing solvated and 18C6 complexed Na+ ion . . . 69 Semilog plots of l/T2.X§' reciprocal temperatures for solutions containing solvated Na+ ion. (x) 0.1 M NaBPh4 in a 20-80 mole % PC-THF mixture; (0) 0.1 M NaBPh4 in a 40-60 mole % MeOH-THF mixture; (+) 0.1 M NaBPh4 in a 60-40 mole % PC-THF mixture . . . . . . . . . . . . . . . . . . 70 Semilog plots of l/T2 XE' reciprocal temperatures for solutions containing 18C6 complexed Na+ ion. (x) 0.1 M NaBPh in a 60-40 mole % PC-THF 4 mixture; (+) 0.1 M NaBPh in a 20-80 4 mole % PC-THF mixture; (0) 0.1 M NaBPh 4 in a 40-60 MeOH-THF mixture . . . . . . . . 71 Computer fits of a sodium-23 nmr spectrum of a solution containing 0.2 M NaBPh4 and 0.1 M 18C6 in THF at 25°C and a delay time of 800 us. (x) eXperimental point; (0) calculated point; (=) no dif- ference between calculated and experimental xiii 12 13 14 15 16 17 18 19 points within plot accuracy. (a) no delay time correction; (b) with delay time correction . . . . . . . . . . . . . . Free induction decays with different relaxation times. . . . . . . . . . . . . . Semilog plots of l/T XE; l/T at various [NaBPh4J/[18C6] mole ratios in tetra- hydrofuran solutions. . . . . . . . . . . . Semilog plot of 1/T XE' 1/T for NaSCN' 18C6 in tetrahydrofuran solutions . . . . . Plot of 1/(T[Na+J ) X§° the inverse total of the free sodium ion concentration for NaBPh4-18C6 in tetrahydrofuran solutions at 25°C . . . . . . . . . . . . . . . . . + . Plots of 1/(T[Na Jtotal) XE' the inverse of the free sodium ion concentration for NaSCN-18C6 in tetrahydrofuran solutions at several temperatures. . . . . . . . . . . . Plot of l/(TENa+] l) XE‘ the inverse tota of the free sodium ion concentration for NaAlEt '18C6 in tetrahydrofuran solutions 4 at 25°C . . . . . . . . . . . . . . . . . . Semilog plot of l/T XE. l/T for NaSCN° 18C6 in methanol solutions. . . . . . . . . Plot of l/(T[Na+] ) vs. the inverse total of the free sodium ion concentration for NaSCN-18C6 in methanol solutions at -4.7°C. xiv 80 81 89 90 93 94 99 101 102 Figure Page 20 Semilog plot of l/(TENa+Jtotal)‘!§' 1/T for NaBPh4°18C6 in propylene carbon- ate solutions . . . . . . . . . . . . . . . 108 21 Plot of l/(TENa+Jtotal)‘!§' the inverse of the free sodium ion concentration for NaBPh4-18C6 in prOpylene carbonate solu- tions at -4.7°C . . . . . . . . . . . . . . 109 22 Semilog plot of k_2 vs. Gutmann donor number for Na+-crown in various solvents. (o) 18C6; (x) DB18C6; (+) DC18C6 (Ref- erences 10, 25, 78) . . . . . . . . . . . . 112 23 Plot of AGi XE' Gutmann donor number for Na+-crown in various solvents. (o) 18C6; (x) DB18C6; (+) DC18C6 (References 10, 25, 78) . . . . . . . . . . . . . . . . 113 24 Semilog plot of l/T gs. l/T for [NaBPh4]/ [18C6] = 1.91 in a 40-60 mole % MeOH-THF mixture . . . . . . . . . . . . . . . . . . 118 25 Plot of l/(T[Na+] ) X§° the inverse total of the free sodium ion concentration for NaBPh ~18C6 in a 40-60 mole % MeOH-THF 4 mixture at 25°C.. . . . . . . . . . . . . . 119 26 Plot of AG?2 XE‘ Gutmann donor number + . . for Na -crown in various solvents (Ref- erences 10, 25, 78) . . . . . . . . . . . . 120 XV Figure 27 28 29 3O 31 32 Page Semilog plots of l/T gs. l/T for NaBPh ~18C6 in 60-40 mole % PC-THF 4 solutions . . . . . . . . . . . . . . . . . 124 Semilog plots of l/T gs. 1/T for NaBPh4'18C6 in 20-80 mole % PC-THF solutions . . . . . . . . . . . . . . . . . 125 Plot of 1/(T[Na+] the total) XE' inverse free sodium ion concentration for NaBPh4'18C6 in 20-80 mole % PC- THF mixtures at 25°C. . . . . . . . . . . . 126 Plot of 1/(T[Na+] ) XE; the total inverse free sodium ion concentration for NaBPh4-18C6 in 60-40 mole % PC- THF mixtures at 25°C. . . . . . . . . . . . 127 Semilog plots of l/T2 XE; 1/T for solvated and complexed cesium ion in methanol solutions. (x) 0.02 M CsSCN; (o) 0.02 g Cs+°DB24C8, SCN', (+) 0.02 g Cs+-DB21C7, SCN’. . . . . . . . . . . . . . 136 Semilog plots of l/T2 XE: l/T for solvated and complexed cesium ion in acetone solutions. (0) 0.02 M CsSCN; (x) 0.02 M CsBPh4 (Reference 30); (o) 0°02.§ Cs+oDB24C8, SCN’; (+) 0.02 M Cs+-DB21C7, SCN'. . . . . . . . . . . . . . 137 xvi Figure 33 34 35 36 37 Page Semilog plots of 1/1 XE- 1/T for CsSCN-DBZ4C8 in methanol solutions. (o)[CsSCN]/[DBZlC7] = 3.18; (x) [CsSCNJ/[DB21C7] = 1.62; (o) [CsSCNJ/[DBZ4C8] = 1.61; (+) [CsSCNJ/[DBZ4C8] = 3.16; all DB24C8 data used for fit . . . . . . . . . . . . 145 Semilog plots of l/T X§° 1/T for CsSCN-DB21C7 and for CsSCN-DB24C8 in acetone solutions. (o)[CsSCN]/[DB24C8] = 1.62; (+)[CsSCN]/[DB24C8[ = 2.55; (x) [CsSCN]/[DB21C7[ = 2.97; (o)[CsSCN]/ [DBZlC7] = 1.62 . . . . . . . . . . . . . 146 Plots of l/(TECs+] ).X§' the total inverse free cesium ion concentration for CsSCN-DBZlC7 in acetone solutions at various temperatures. (0) 215°K; (X) 200°K . . . . . . . . . . . . . . . . 147 Plots of l/(T[CS+] the total) Xé' inverse free cesium ion concentration for CsSCN-DB21C7 in methanol solutions at various temperatures. (0) 220°K; (X) 200°K . . . . . . . . . . . . . . . . 148 Plots of 1/ (T[CS+] the total) XE‘ inverse free cesium ion concentration xvii Figure 38 Page for CsSCN°DB24C8 in acetone solutions at various temperatures. (x) 215°K; (o) 200°K . . . . . . . . . . . . . . . . 149 Plots of l/(T[Cs+] ) XE' the total inverse free cesium ion concentration for CsSCN-DB24C8 in methanol solutions at various temperatures. (0) 2140K; (x) 206°K . . . . . . . . . . . . . . . . 150 xviii THF PC MeOH EtOH DMF DME LIST OF ABBREVIATIONS Tetrahydrofuran Propylene carbonate Methanol Ethanol Dimethylformamide Dimethoxyethane xix CHAPTER 1 HISTORICAL REVIEW INTRODUCTION The abundance of information available in the field of alkali metal ion complexation has grown dramatically since the discovery by Pedersen (1,2) of the macrocyclic ligands known as crown ethers and later the synthesis of cryptands by Lehn and coworkers (3-5). Yet, the knowledge of com- plexation kinetics of alkali metal ions with crown ethers is meager compared to what is currently known about com- plex stabilities. This is surprising considering the im- portance such information may provide as, for example, models for ion transport in biological membranes. The role of the solvent in the kinetic process of com- plexation of alkali metal ions by crown ethers is, as yet, unknown. The influence of the ligand cavity size and of the counterion on the complexation kinetics remain unex- plained. Whether or not the different alkali metal ions exhibit differences in their complexation kinetics with crown ethers has not been established. These are just a few of the unknowns which exist in this area. It is the intent of this dissertation to begin the process of unraveling these unknowns. A review of current progress in the field of kinetics of complexation of alkali metal ion by crown ethers is necessary at this point. COMPLEXATION KINETICS OF ALKALI CATION - CROWN ETHER COMPLEXES The sparse information available for the kinetics of alkali metal ion complexation by macrocyclic ligands has been reviewed by Liesegang and Eyrins (6) (through 1978), by Schmidt (7) (through 1981) and by Szczygiel (8) (through 1984). Consequently, this survey will focus on the main concepts of alkali metal ion - crown ether kinetics dis- cussed in these reviews, as well as review current progress in this field. Several mechanisms have been postulated or observed for the complexation of alkali metal ions by crown ethers and cryptands (Figure 1). These mechanisms are listed in Table l. The presence of the solvent, though not indi- cated, is implied. Wong 33 2;. (9) used proton nmr to study the exchange of sodium ion with dimethyldibenzo-l8-crown-6 in deuterated tetrahydrofuran. Since an excess of the ligand was pres- ent in solution, exchange mechanism I was assumed in which the metal ion is simply exchanging between the crown ether molecules. An Arrhenius activation energy, Ea, of 12.5 kcal-mol—l was calculated for this process. Replacement of the above crown by dicyclohexyl-lB-crown-6 (DC18C6) resulted in a drop in the coalesence temperature from 2°C. In fact the coalesence was not observed although the measurements were carried down to -60°C. The difference 18-Crown—6 12—Crown—4 Cryptand C211 0 W; it) 5 NH 5 VALINOMYCIN j_ 3 My; "Kiwkgo o \ R‘ .n’m’m‘mn, nomm A a‘ a R2 a R3 a CH, n‘ , czus MONACTRN _3_ R' a 51’: CH, R2 = a‘: c2145 DINACTIN 5. R‘ . CH, 9’ a R3 = n‘ = c2145 TRINACTIN _5_ R‘ = R2: R’. R‘ : CZHS TETRANACTIN _6_ Some synthetic and naturally occurring macro- Figure 1. cycles. Table 1. Exchange Mechanisms Postulated or Observed for Alkali Metal Ion-Crown Ether Complexation.’ k 'k M+°C + c—i M+-C + c I k M+ + C —L M+°C II T— k--1 M+ + M+-C—]-(-l- M+ c + 34* III @- ° C. k_ 4+ k k 0c 0+M+_.__:_L_S.M+'C Iv IE-1 k k M+ + c—LM+-C—3—\-M+c' v k k k + 1 -+ 2 -+ 3 + M+c———--M...c———-MC_——>(MC) VI E-1 E-2 i2-3 in kinetic properties of the two crowns was postulated to be due to stronger ion-pairing of the dimethyldibenzo-18- crown-6 complex with the florenyl counterion which was used. Shchori SE.E$° (10) used 23 Na nmr to study the com- plexation kinetics of Na+ by dibenzo-18C6 (DB18C6) in di- methylformamide (DMF). They found that in this case the exchange occurs by way of the associative-dissociative mechanism II rather than the bimolecular process III. Chock (11) used temperature jump relaxation technique to study monovalent cation complexation by dibenzo-30- crown-10 in methanol (MeOH) solutions. The kinetic data were best described by mechanism IV. The crown ether undergoes a rapid transformation between two confomers. One of the confomers reacts readily with the cation to form the complex. The conformational equilibria of the smaller crown ethers have been explored further. Liesegang e3 21. (12) measured relaxation of 18C6 in aqueous solutions with no salt present and found a relaxation time of I_1 = 6.2 x 108 s—1 at 25°C. A relaxation rate of T_l = 1.4 x 108 s-1 was measured for aqueous solutions of 15C5 by Rodriquez .23.31’ (13). In both cases the process was assumed to be a conformational change C'-%3L-C. The longer relaxation -0 process for 15C5 as compared to 18C6 is consistent with a lower flexibility of the smaller crown. The relaxation of 18C6 has also been measured in several other solvents and the kinetic data are summarized in Table 2. With the exception of water, the rate con- stants at 25°C have been calculated assuming AHI and A875 are independent of the temperature. It may readily be seen that there is a large dependence of ARI and AS? on the sol- vent. However, at 25°C with the exception of ethanol, k__0 is always ~107 s"1 at 25°C, and with the exception of water, k0 is always-v2 x 109 5-1. In the two exceptional cases the rate constants differ from the others by less than one order of magnitude. It is interesting to note that Chen and Petrucci (15) report that in methanol solutions one mole- cule of the solvent is eliminated during the isomerization process. Presumably, the open form of the crown is "sol- vated" or "complexed" with the methanol and releases one molecule during the conformational change. The importance of such crown - solvent interactions has been demonstrated by P. Boss (18). Based on the observation of crown conformational equi- libria in the absence of metal ions, one might expect similar conformational equilibria in the case of the crown - metal ion cOmplex, i.e., mechanism V shown in Table 1. This mechanism fits best the results of Grell gg‘gl. (19) for the complexation of M1 by the naturally occurring anti- biotic valinomycin in methanol solutions. The first step involves a rapid diffusion-controlled collision between the ion and the valinomycin molecule which is followed by and-ms HIHoE.Hmoxm .Humu .ha mocmummmmm .wa mocoummmmp .mH mocmummwmo .va mocmnmmmm n .ma mocmummmmm oaXo.N8 m xooomuvxncmcmxoflo mOme.H moaxm.m AUomNV o w.v m.mNI H.H mOHNm.v wOHXm.H AUomHIVAmvmoam mOHxH.N mcaxo.m AUova vm.m m.m c.0Hl N.m mOHxH.m woaxo.m8 AUooNIVAUvamE moaXm.m oOHXm.m AUova . . .- . . x. - N CH m h m ma w m moaxm m moa o m AUOOH VACVMSD . . . . . . N h h v N h h N CH mOHxN m N.o._uu_no H AUommvava m o 0 0| 0| 0 ol Ancxms lmcxms lgcxms Imcxms luv x Inc I .mucm>aom msoHum> ca mUmH mo :oflummemm How mumumfimnmm Daumcflx .m OHQMB a rate limiting conformational rearrangement of the val- inomycin to form a compact complex. The first observation of a conformational rearrangement upon complexation for a crown ether was made by Chen and Petrucci (15) for the sodium ion with 18C6 in methanol using ultrasonic absorption. They determined a forward rate con- stant for this process to be k2 = 2.8 x 108 5.1. This is a slower process than the conformational change of the crown ether in the absence of sodium (k0 = 2.1 x 109 s-l) This is not surprising since the complex should be less flexible than the crown ether alone. In two interesting publications Petrucci and coworkers (16,17) have postulated a third conformational arrangement of the 18C6 complex in ethanol and in DMF. They found that in some instances mechanism VI is applicable instead of mechanism V. The authors define MI...C as a solvent separated pair, MC+ as a contact pair in which there is some coordination by the crown and (MC)+ as an included species in which the ion is fully encapsulated by the crown. Two concentration independent relaxation processes were observed for the systems NaClO4-18C6 and KSCN-18C6 in DMF and LiClO4-18C6 in ethanol solutions instead of the single relaxations observed for NaClO4-18C6 and for KClO4-18C6 in ethanol and for LiClO4-18C6 in DMF. Thus, the authors con- cluded that the solvent strongly influences the complexation 10 kinetics. The first step of the complexation process was assumed to be a partial desolvation of the ion and ligand rearrangement. The second step is the encapsulation of the metal ion with the subsequent rearrangement of the crown. The solvent influence was presumed to arise in the second step of complexation. At this point, cation de- solvation should be complete in order for the 18C6 to en— capsulate the ion. The observation of two relaxation pro- cesses for K+ and for Na+ with 18C6 in DMF but only one in ethanol is attributed to solvent differences. Chen and Petrucci concluded that the dielectric constants of the two solvents are not responsible for the observed solvent ef- fect. They attributed the solvent influence to be due to differences in the solvating abilities of the solvent. The authors postulated that in the case of Na+ and K+ complexa- in DMF the strong solvation of the ions sufficiently slows down the second step of complexation to make it observable for these ions. They speculate that the Li+ ion is so strongly solvated that this step does not occur to an ap- preciable extent and thus, is not observed. In ethanol, on the other hand, the metal ions are pre- sumed to be less strongly solvated. Thus, the second step of complexation is too rapid to be observed for K+ and for + complexation. With Li+ the desolvation step is suf- Na ficiently fast to be observed. It should be mentioned that the Gutmann donor numbers (20) for ethanol and DMF 11 have been determined (20,21) to be 31.5 and 26.6, respec- tively. The Gutmann donor number is defined as the negative enthalpy upon complexation of the solvents with SbCl5 in 1,2-dichloroethane, 142:1 1,2 DCE S + SbCl5 —————9 S'SbCl5 D.N. = -AH Therefore, according to this solvent donicity scale ethanol is a better solvator of the sodium ion than is DMF. This is in contrast to the conclusions just described. The first direct conformational rearrangement of the 18C6 complex by nmr was reported by Dickert and Bumbrecht (22). They observed two conformers of the Co2+ complex by proton nmr at -l°C in CD3NO2 with small amounts of MeOH present. These two conformers were postulated to be the two species Co (18C6)2+ and mer-[Co(18C6)(MeOH)3]2+. Indirect nmr evidence of conformational rearrangement upon complexation has been observed by Mei gg‘al. (23) in their 133 Cs studies with the cryptand C222-Cs complex in acetone (AC) and in propylene carbonate (PC) solutions. Their results of the exchange rate and chemical shift study suggested the existence of the following equilibria: Cs+ + c222 ——\ (Cs'C222)+ ——>- (CS°C222)'!' ‘?*' ex eom Amy ACV Anvx AOVK Anvx ADV M.UomN um mucw>aom msoHum> ca oOmH cues coflumxwameou coH +x How mumqumumm oaumcex .v mange 16 The authors have suggested the large variation may be due to differences in solvation of K+ in the transition state. The weaker solvents show larger activation energies because of the charge-charge repulsion which exists in the transition state is greater. Mei 23 31. (28,29) studied Cs+ ion complexation kinetics with 18C6 and DC18C6 in pyridine and in PC solutions by 133Cs nmr. The results are listed in Table 5. The associa- tive-dissociative mechanism was assumed to be operative. It is interesting to note that the rate data are essentially the same for the two 18C6 analogues despite the difference in solvents. The activation energy values are approximately the same as those found by Shchori 33 21' (25) for the Na+- DB18C6 exchange in methanol solutions. However, the activa- tion energies are lower than those for the K+-18C6 complex. 133Cs nmr to investigate Cs+ Shamsipur (30) employed complexation kinetics with the larger crown ethers DB21C7, DBZ4C8, and DB30C10 in methanol and in acetone. The re- sults are listed in Table 6. The associative-dissociative mechanism was assumed to apply. In all cases, the activa- tion energy for the decomplexation step was found to be larger in acetone than in methanol. The solvent dependence of the activation energy increases with decreasing cavity size of the ligand. The activation entrOpy, however, is more negative in methanol than in acetone with the result that in both solvents the free energy of activation, AGfl, 17 Table 5. Kinetic Parameters for Complexation Kinetics of Cs+ Ion at 25°C.a Sol. Crown 10—3k_§b) Eéc) AH#(C) AS#(d) AG#(C) pc DC18C6 11. 8.5 7.9 -14 11.94 py 18C6 9.5 8.4 7.8 -14.2 12.03 aReferences 28, 29. bs-l. Ckcal-mol-l d e.u. 18 H IHOE.Hmox 0:. $0 U . m HI 0 .Hum Huse .om mocmummmm CH pcsom Ho Eoum pODMHsono muasmmmm m.m N.HNI m.ol v.HH b.5HI H.m h.m _vOHXm.N mOHXN.m mow: hUHNmO v.m H.mNI «.ml m.OH m.h I h.m m.m HOme.m wOHXH.h Dd hUHNmQ h.m m.NmI H.¢I 5.0H m.oHI h.m m.© vOHxh.w wOHxv.v mom: mU¢NmD v.m m.omI m.mI m.OH H.0HI 0.5 N.m vOme.OH mOHXh.o 04 mU¢NmD v.m H.mvl oo.ml N.HH m.NNI o.v N.m vOHXm.m mOme.h mow: OHUome m.m m.OVI oo.ml m.HH m.omI m.m H.w @OHXH.N mOon.m U< OHUome H H H HI HI HI Hl.m HI H . lessee lmcxme levies lecxoe Aocxme lecxme 163 e 163 s lee I How 63666 m.Uomm um mCOHusHOm wcoumom :H paw Hocmnumz CH muonum czonu Hmum>wm nufi3 coH +mu mo cofiumxmameou How mumumEmumm UHumcwx .o magma 19 is essentially the same for the three crown ether com- plexes. It is interesting to note that the activation energy for the complexation step is negative, i.e., the transition state is enthalpically more stable than the reactants. However, the large decrease in entropy results in a positive free energy of activation for the forward reaction. In general, the cation exchange between free and com- plexed sites at room temperature is fast on the nmr time scale, and only one population-averaged nmr signal of the alkali metal nucleus is observed when the M+ concentration is greater than that of a crown ether. However, Lin and Popov (31) have recently reported a slow exchange at room temperature for the system NaBPh4-18C6 in tetrahydrofuran and in 1,3-dioxalane solutions. But, when the counter- ion was ClOZ or I- the exchange was fast. Although the kinetic parameters for these systems were not measured, there are two interesting observations which may be made from this study. First, this is the first known observation of slow exchange on the nmr timescale for an alkali metal ion with a crown ether at room temperature. Second, this is the first known occurrence of an anion influence on the exchange process. In conclusion, while work is accelerating in the field of complexation kinetics of alkali metal ions with crown ethers a great deal of study is still required to 20 understand fully these processes in terms of solvent, cation, and anion influences to name a few. CHAPTER 2 EXPERIMENTAL PART 21 A. Salt and Ligand Purification Sodium tetraphenylborate (Aldrich, Gold Label) was dried under vacuum at room temperature for 48 hours. Re- agent grade sodium thiocyanate (Mallinckrodt) was re- crystallized from acetonitrile and dried under vacuum at 60°C for at least one day. Reagent grade sodium perchlor- ate and sodium iodide (Matheson Coleman & Bell) were dried at 110°C for several days. Cesium thiocyanate (Pfaltz and Bauer) was dried under vacuum at 60°C for at least two days. Sodium tetraethylaluminate was originally donated by M. C. Day (32). Additional amounts of the salt were synthe- sized by the method of Hohn 3E El“ (33). All glassware was cleaned and dried thoroughly. The reaction and all handl- ing of materials were carried out under dry nitrogen atmosphere. The reaction for the preparation of NaAlEt4 is: reflux 3Na + 4A1Et3-—————%> 3NaA1Et4 + Al toluene Forty ml of AlEt3 were added by means of a dropping funnel to a refluxing solution containing 15 g of sodium metal in toluene. The solution was allowed to reflux for 22 23 two h after which time a hot filtration was carried out in a glove box yielding a clear solution. Upon cooling a white product precipitated which was the desired compound. The NaAlEt4 was then recrystallized from toluene and stored under vacuum. All sample preparations involving this salt were done under a nitrogen atmosphere in a glove box. The melting point of the salt was 110-127°C (literature m.p. = 125°C (33)). It is likely that the solvent was not com- pletely removed from the salt thus extending its melting point range. Day and coworkers (34) recommend rinsing the salt with hexane in order to help reduce trace amounts of toluene. This was not done since the product was only to be used for qualitative tests. Tetrahydrofuran (Mallinckrodt) was refluxed over a mix- ture of potassium metal and benzophenone for at least 24 h. Methanol (Fisher) was refluxed over sodium methoxide for one day. Propylene carbonate (Aldrich) was refluxed over barium oxide under reduced pressure for at least one day. Acetone (Fisher) was refluxed over calcium sulfate for two days. In all distillations, only the middle 60 percent of the solvent fractions were kept. The solvents were stored over activated molecular sieves in a dry box under nitrogen atmosphere. Water content of the solvents was always less than 100 ppm as determined by gas chrom- atography. The macrocyclic polyether 18-crown-6 (Aldrich) was 24 recrystallized from acetonitrile (35) and dried under vacuum for two days at room temperature. The purified ligand melted at 37-38°C (lit. m.p. 36.5-38.0°C (36), 38- 40°C (37)). The crown ethers dibenzo-21-crown-7 and dibenzo- 24-crown-8 (Parish) were recrystallized from n-heptane and dried under vacuum at room temperature for at least two days. The melting points were 107.0°C and ll3.0°C (38), respectively, which are the same as the reported values. B. Instrumental Measurements and Data Handling 1. Instruments. Sodium-23 nmr measurements were obtain- ed on a Bruker WH-180 spectrometer operating at a field of 42.3 kG and a frequency of 47.61 MHz and on a modified (39) DA-60 spectrometer operating at a field of 14.09 kG and a frequency of 15.87 MHz. Both instruments were operated in the pulsed Fourier transform mode. A Nicolet 1180 computer was used to carry out data manipulation on the WH-180 while a Nicolet 1080 computer was used on the DA-60. Cesium-133 nmr measurements were obtained on a Bruker WH-180 spectrometer operating at a field of 42.3 kG and a frequency of 23.62 MHz. 2. References and Corrections. The reported chemical shifts are referenced to infinitely dilute aqueous sodium chloride (sodium-23 nmr measurements) or to infinitely dilute aqueous cesium chloride (cesium-133 measurements). 25 Downfield chemical shifts are taken to be positive. The chemical shifts are corrected for differences in bulk diamagnetic susceptibility between samples and reference solvent according to the equations of Martin gg'gl. (40). For a Fourier transform experiment utilizing an electro- magnetic (DA-60): 6corr — sobs 7T (Xref Xsamp) where xref and Xsamp are the unitless volumetric suscepti- bilities of the reference and solvents, respectively, and 6corr and sobs are the corrected and observed chemical shifts, respectively. For a Fourier transform experiment utilizing a super- conducting magnet (WH-180): where the symbols have the same meanings as above. It was shown by Templeman and Van Geet (41) that at low salt con- centrations the contribution of the salt to the volumetric susceptibility of the solutions can be neglected. When mixed solvents were used the volumetric susceptibility of the solutions were calculated as follows (42): 26 calc where XAB = volumetric susceptibility of the solution, VA = volume of solvent A, VB = volume of solvent B, HA = volume diamagnetic susceptibilities of pure solvent A, and XB = volume susceptibility of pure solvent B. Two methods were employed for variable temperature ex- periments, depending on the instrument used, to reference all chemical shifts to 25°C. When the WH-180 was used the sample was placed in a 10 mm O.D. tube with a 5 mm tube containing lock solvent mounted coaxially inside. The sample was placed in the magnet and the lock maintained throughout the variable temperature experiment. At each temperature, only the fine Z shim control was adjusted to maintain lock and field homogeneity. Because the field was locked throughout the experiment chemical shifts at each temperature could be referenced to the one at 25°C. Finally, the chemical shifts could all be referenced to infinite dilution at 25°C by taking the spectrum of the sample with the appropriate reference at 25°C. Chemical shift measurements as a function of temperature were much easier to do on the DA-60 spectrometer. This instrument has an external lock system which allows re- placement of the sample with the reference solution while maintaining lock. The reference solution was placed in a 5 mm O.D. tube which was then vacuum sealed in a 10 mm O.D. tube as reported by Mei 33 El; (29). They showed this arrangement to be quite convenient for keeping the 27 reference solution at room temperature while inside the magnet for short periods of time. Thus, all chemical shift measurements determined in this manner are referenced to that at 25°C. 3. Temperature Calibration and Control. Temperature was controlled on the WH-180 with a Bruker B-ST 100/700 tempera- ture control unit and measured to 21°C with a calibrated Doric digital thermocouple placed about 1 cm below the sam- ple. A N2 flow rate of 50 SCFM was used and the tempera- ture calibrated by means of the chemical shift difference between the methyl and hydroxyl protons of methanol (43). The data are listed in Table 7 and the resulting calibration curve was used to determine the sample temperature. On the Varian DA-60 spectrometer, temperature was con- trolled with a Varian V-4540 temperature control. Sample temperature was measured with a calibrated Doric digital thermocouple placed about 1 cm below the sample. A N2 flow rate of 40 SCFM was used. The Doric unit was calibrated by placing a calibrated thermometer in a 10 mm sample tube containing methanol and taking thermocouple and thermometer readings at various temperatures. The data are given in Table 8. Due to the close agreement between thermocouple and thermometer values, it was unnecessary to use a cali- bration curve. 28 Table 7. Temperature Calibration of the WH-180 Spectrometer.a Temperature Read (°C) Actual Temperature (°C) 54.9 55.0(i1) 44.3 44.8 34.6 35.8 24.0 24.4 13.5 14.4 3.9 3.6 -5.7 -5.2 -15.2 -16.2 -25.0 -27.2 -35.6 -38.4 -45.7 -48.4 a20 mm high frequency probe. 29 Table 8. Temperature Calibration of the DA-60 Spec- trometer. Temperature Read (°C) Actual Temperature (°C) 47.3 47.0 36.9 36.6 23.3 23.4 12.6 13.0 2.5 3.0 -13.4 -13.5 -18.1 -18.0 -23.8 -23.5 30 4. Linewidth Measurements. Linewidth measurements on the WH-180 were done by either measuring the width at half height or by fitting the Spectra to a Lorentzian function. On the DA-60 spectrometer, linewidths were measured by plotting the spectra and measuring the width at half height. In order to correct the linewidths for daily differences in field homogeneity, two techniques were used, depending on the nmr spectrometer. 0n the WH-180 field homogeneity was checked by comparing the linewidth of a sample with that obtained on a previous day. In general, the linewidth was the same within experimental error. Total error in line- width measurements (and thus, l/TZ's since l/(nTZ) = Avg) is estimated to be 10% of the measured value. Because the DA-60 spectrometer has an external lock it is possible to replace the sample with a reference solution as described in Section 2. Besides its use as a chemical shift reference, an aqueous 3 M NaCl solution was used as a linewidth reference. The observed linewidth of a nucleus can be viewed as the sum of the natural linewidth and the contribution from field inhomogeneity. Thus, 1 l l (I?) = (T—) + (T—) 2 observed 2 natural 2 inhomogeneity 23 In order to determine the natural Na linewidth for the aqueous 3 M NaCl reference the following procedure was per- 7 formed. First, the Li linewidth of an aqueous 4 M LiClO4 31 solution was measured and found to be 1.56 Hz. The natural linewidth should be almost 0.0 Hz (44). Thus, the inhomo- geneity contribution to the linewidth is 1.56 Hz. Next, the sodium-23 nmr of the 3‘M NaCl solution was taken. Care was taken to ensure the geometry was the same as with the LiClO4 solution. The field was kept locked between samples and there was no need to remove the probe. A 23Na linewidth of 7.68 Hz was measured. Thus, for 3 M NaCl ) = (7.68-l.56)n = 19.2 5’1 natural or Av;i = 6.12 Hz. Eisenstadt and Friedman (45) found a concentration depen- 23 dence of the Na relaxation for aqueous NaCl solutions. The relationship has the form: 1 _ T; — 17.55 + (0.55) CNaCl where CNaCl = molarity NaCl. Though their maximum concen- tration was only 1 M NaCl, if an extrapolation is made to 3 M NaCl a first approximation of the relaxation rate may be obtained. A value of 1/T2 = 19.2 s-1 (or Avg = 6.11 Hz) is calculated, which is the same as that measured above. 23 Thus, all Na linewidths were referred to aqueous 3 M NaCl which was also used as a chemical shift reference. 32 Typically, corrections in linewidths were on the order of 1-3 Hz. Total error in these measurements is estimated to be 10% of the measured value. Differences in sample susceptibilities (due to different solvents) will affect field homogeneities to different ex- tents. Thus, the corrections made for field inhomogeneity may not yield true (l/TZ) values for a particular sample. However, the referencing procedure described above allows one to refer the linewidths to a common value and, thus, maintain consistency on a day to day basis. 5. Data Acquisition and Signal Processing. Several instrumental techniques were used to increase the signal to noise ratio and digital resolution. To increase signal/ noise, two techniques were used: signal averaging and ex- ponential line broadening of the free induction decay (FID). Typically, loo—10,000 scans were collected, depend- ing on sample concentration and linewidth, and linebroaden- ing of 10-40 Hz was applied to the FID. To increase digital resolution the zero filling tech- nique (46) was employed. In general, one collects an N- point FID and then adds 2—4 N zeroes to the end of the FID before Fourier transformation. This allows an increase in the point to point resolution without distorting the lineshape. 33 6. Data Transfer. For kinetic measurements it was desired to obtain theoretical fits of nmr lineshapes to extract the kinetic information present. The method of data transfer varied depending on the nmr instrument used, but ultimately was processed on a CDC-6500 mainframe com- puter. 6.a. WH-180. Data transfer from the WH-180 to the CDC-6500 occurred in 3 distinct stages. First, the transformed FID was stored on the hard disk used by the Nicolet 1180 computer. The data were transferred to a Digital Equipments Corporation (DEC) PDP-ll computer system via two programs. A program, MOVE, based on a pro- gram written by Walmsley and Atkinson (47), was written by K. Johnson and stored along with the computer language, BASIC, on the disk containing data to be transferred. The purpose of MOVE is to send the data out Serial Port B of the Nicolet 1180 computer. With the software update on the WH-180 in 1984 a new program, NTCDTL, was used to send data out this port of the computer. NTCDTL was provided by Nicolet with the new software package. A copy of MOVE as well as examples of the use of this program and of NTCDTL can be found in Appendix I. A program CDC (48), Operating on the DEC system, takes the data being transferred and stores the data on a floppy disk. The second stage of data transfer involves taking the 34 data on the floppies and reformating them to KINFIT (49) acceptable format. KINFIT is a nonlinear least squares fitting routine used for data analysis. The program NIC180 was used to reformat the data transferred by MOVE. NIC84 was used to reformat data transferred by NTCDTL. Both NIC180 and NICDTL were written by T. V. Atkinson (49). The third stage consists of transcribing the new data files onto a magnetic tape and then transferring the data from the tape to the CDC-6500 computer. The files may then be called by KINFIT. 6.b. 25:20. Data to be computer fit was plotted to obtain a hard c0py. The plots were digitized using a Science Accessories Corporation GP-3 digitizer available at the Physiology department. The digitized data were then punched onto computer cards in KINFIT acceptable for- mat and run on the CDC-6500 computer. CHAPTER 3 MISCELLANEOUS 35 A. ION-PAIRING IN TETRAHYDROFURAN SOLUTIONS INTRODUCTION It is reasonable to assume that the observation of an anion influence on the complexation kinetics of sodium ion with 18C6 in tetrahydrofuran, as discussed in Chapter 4, is due to differences in ionic association of the sodium salts and of their complexes; especially since THF has a low di- electric constant (D = 7.6). The salt which exhibits slow exchange at room temperature (NaBPh4) has a large, "soft", bulky anion while the fast exchange salts (NaSCN, NaClO4, and NaI) have small, "hard" counterions. Previous workers (50-55), using conductance measure- ments, have concluded that in tetrahydrofuran solutions, NaBPh4 exists primarily in the form of solvent separated ion-pairs (Figure 2). Recently, Chabanel EE.El° (56), using infrared spectrosc0py, have concluded that NaSCN exists in the form of contact ion-pairs as well as higher ionic aggregates in THF (Figure 2). Greenberg and coworkers (50), using sodium-23 nmr, observed a concentration independent chemical shift in THF when BPh4- was the counterion. How- ever, SCN-, I-, and C10 - all influenced the sodium-23 4 chemical shift. Thus, contact ion-pairing is significant for the fast exchange anions but not for BPh4 ion. How- ever, no known studies have been published concerning the ion-pairing of complexed sodium salts in THF. 36 37 s SIvI‘s x“ s Solvent-separated s 8 NW 8 Contact Figure 2. Types of ion pairs in solution. Note: most experimental techniques can discern only those solvent-separated ion pairs which are separated by a single solvent molecule. 38 RESULTS AND DISCUSSION 1. Conductance Measurements The equivalent conductances of 0.1 M sodium salt solu- tions and of their complexes with 18C6 (when soluble at this concentration) were determined and are listed in Table 9. Also listed is the conductance of 0.1 M NaAlEt4 in THF reported by Day and coworkers (57). The results for the uncomplexed sodium salt solutions will be considered first. The conductances of the NaBPh4 and the NaAlEt4 solu- tions are two orders of magnitude greater than those of the "fast" exchange salts. As mentioned above, NaBPh4 is be- lieved to be predominantly in the form of solvent separated ion-pairs whereas NaSCN, NaI, and NaClO4 are presumed to form contact ion-pairs and higher ionic aggregates in THF. The conductance data seem to agree with these conclusions since highly associated salts would be poor conductors. Day and coworkers (58) have concluded that NaAlEt4 exists primarily in the form of solvent separated ion-pairs in THF. The conductance data also suggests that this salt more closely resembles NaBPh4 (i.e., solvent separated ion- pairs), than the other salts in tetrahydrofuran. The conductance of the complexed NaBPh4 salt solution has a slightly lower conductance than the pure salt solu- tion. If strong contact ion-pairing is minimal, this would be the expected behavior since the mobility of the sodium 39 Table 9. Equivalent Conductances of Various Tetrahydro- furan Solutions at 25°C. A (mho-cmz-equiv.-l) Solution 398 Hz 629 Hz 971 Hz 1942 Hz 3876 Hz 0.09999M NaBPh4 20.98 20.86 20.75 20.64 20.44 0-114 NaAlEt4 -------------------- 24.00‘3) 0.1004M NaSCN 0.1365 0.1430 0.1420 0 1414 0.1381 0.09979M NaClO4 0.4263 0.4465 0.4436 ------------ 0.1001M NaI 0.2203 0.2187 0.2169 0.2151 ------ 0.1001MNaBPh4 + 0.1108M 18C6 15.04 15.00 14.92 14.83 14.68 0.09990M NaSCN + 0.1123M.18C6 1.942 2.035 2.022 2.017 1.968 aReference 58; @3000 Hz. 40 ion would be reduced when complexed. Thus, the NaBPh4- 18C6 complex exists primarily in the form of a crown separated ion-pair. When NaSCN is complexed by 18C6 the conductance actually increases as compared to the solution of pure NaSCN. It is likely that this is due to the breakup of higher ionic aggregates which are found in the uncomplexed NaSCN solu- tion. The conductance of the complexed NaSCN solution is. still one order of magnitude lower than that of the com- plexed NaBPh4 solution. Mobility considerations would argue in favor of the complexed NaBPh4 having the lower conductivity if ion-pairing is negligible since BPh4- is the larger counterion. Thus, the complexed NaSCN must be contact ion-paired to some extent. 2. Infrared Studies The room temperature infrared spectra of the C-N stretch- ing region of the SCN- ion for a 0.1 M NaSCN solution and of its complex with 18C6 in THF are shown in Figure 3. The spectrum of the uncomplexed salt agrees well with that re- ported by Chabanel 3E 21. (56). Their band assignments are those listed. It is important to note the absence of a 1 band at 2052 cm- which would correspond to "free" SCN- ion (56,59,60). Upon complexation with 18C6 a single band at 2058 cm-1 remains. This band was observed in the pure salt solution Arbitrary TronsmiIIonce Figure 3. 41 I0)” (c) / l I 2|OO 2050 2000 27(cm") Room temperature infrared spectra of tetrahydro- furan solutions containing NaSCN. I) 0.126 M NaSCN + 0.140 M 18C6: Spectrum of Na+-18C6-NES'. II) 0-074.fl NESCN, (a) Na+-Ncs’-Na+; (b) Na+- NCS'; (c) (Na+-NCS')2. Assignments from Ref- erence 56. 42 and corresponds to the contact ion-paired SCN- ion. Thus, the complexed NaSCN is contact ion-paired. No free SCN- band is observed for this solution. As expected from the conductance results, the complexation of the sodium ion has eliminated the higher ionic aggregates found in the pure salt solution. Thus, the conductance and infrared results complement each other. 3. NaAlEtA-18C6 Exchange in THF The observation of similar ionic states for NaBPh4 and NaAlEt4 in THF solutions suggests that the system NaAlEt4- 18C6 may also exhibit slow exchange at room temperature on the sodium-23 nmr timescale in tetrahydrofuran. Approxi- mately a 2:1 mole ratio of NaAlEt4 to 18C6 solution was prepared. The room temperature sodium-23 nmr spectrum is shown in Figure 4. As can be seen, this salt also exhibits slow exchange with 18C6 in THF. Thus, BPh4- is not unusual in being the only anion to exhibit slow exchange at room temperature. Furthermore, the belief is strengthened that differencesimlionic association are responsible for the anion influence on the sodium ion-18C6 exchange rate. CONCLUSIONS In summary, the slow exchange salts and their complexes with 18C6 form predominantly solvent separated (or crown Temp. (°C) 49.2 41.6 25.3 Figure 4. Sodium-23 nmr of a solution containing [NaAlEt4J/ [18C6] zl.9 in THF at several temperatures. 44 separated) ion-pairs in tetrahydrofuran solutions. The fast exchange salts and their complexes with 18C6 form pre- dominantly contact ion-pairs or higher ionic aggregates. 45 3B. ISOSOLVATION STUDIES INTRODUCTION The role of the solvent has long been recognized to be important in influencing solution equilibria, ionic inter- actions, reaction kinetics, etc. For example, the stability constant of a complex can change by several orders of mag- nitude simply by replacing one solvent medium by another. In a "simple" reaction of a solvated metal ion forming a complex with a ligand the solvent may be viewed as a com- peting ligand. Thus, the reaction should be written: + ——>- + M°Sn+LV-—M-L+DS where MI is the metal ion, L is the ligand, and there are n solvent molecules, S, solvating the metal ion. It is reasonable to assume that the strength of the solvent-ion interactions are important in determining the above equi- librium (solvent interaction with the ligand can also be important). The need to compare solvating abilities of a given series of solvents has produced several solvent rating systems including the Gutmann donor number scale (20). This "donicity" scale has been found to correlate quite remark- ably with the infinite dilution sodium-23 nmr chemical shifts in the solvents (61). 46 The concept of preferential solvation of ions in sol- vent mixtures is an additional technique which has been used to compare relative solvating abilities of solvents. Several reviews (62,63) on the topic currently exist and therefore, only the general concept and assumptions will be presented here. Frankel 32 £1. (64) studied Co(acac)3 solvation in chloroform-CCl4 mixtures by following the cobalt-59 chemi- cal shift as the solvent composition was varied. They prOposed that a nonlinear relationship between this param- eter vs. solvent composition is an indication of preferen- tial solvation. The "isosolvation point" was defined as the composition at which the chemical shift of the solute lies midway between the values in the pure solvents. It was postulated that at this composition the contributions from the two solvents to the solvation sphere of the ion are equivalent. If the two solvents have unequal solvating abilities then the solvent with the lower composition at the isosolvation point is the stronger solvator. Popov SE.21° (42,65) have studied the Na+ ion preferen- 23Na nmr. tial solvation in a variety of solvent mixtures by In general, the solvent which showed preferential solvation for sodium ion in a mixture had the higher Gutmann donor number. Among the more interesting results, however, it was concluded that the Na+ ion is preferentially solvated by dimethylsulfoxide (DMSO) in mixtures of DMSO and 47 pyridine. This was unexpected since pyridine has a higher Gutmann donor number than DMSO. The phenomenon was explain- ed as due to the destruction of the associated structure of DMSO by the addition of pyridine which thus enhanced its solvating ability (66). The authors, therefore, cautioned that the properties of solvents in a solvent mixture may be quite unexpected considering those of the neat solvents. In summary, the study of mixed solvent systems can be very useful in the comparison of relative solvating abili- ties of a given pair of solvents. In addition, such studies may uncover those solvent mixtures which exhibit unexpected properties as compared to the pure solvents. We have used sodium-23 nmr to study the preferential solvation of and NaClO sodium ion for the salts NaBPh in the systems 4 4 methanol-tetrahydrofuran and propylene carbonate-tetrahydro- furan. RESULTS AND DISCUSSION The sodium-23 nmr chemical shifts as a function of sol- vent composition have been determined for 0.1 M salt solu- tions of NaBPh4 and of NaClO4 in PC-THF and in MeOH-THF mixtures. The results are tabulated in Tables 10 and 11. The data are shown graphically in Figures 5 and 6. The "isosolvation point", defined on page 7, has been determined for all systems with the exception of NaBPh4 in MeOH-THF mixtures (NaBPh4 is not sufficiently soluble 48 Table 10. Sodium-23 Chemical Shift vs. Composition of THF/PC Binary Mixtures at 25°C. 6Na+(a) XTHF NaBPh4(b’C) NaClO4(b'd) 1.000 -7.61 (i0.10) -8.39 ($0.10) 0.793 -8.04 -8.26 0.663 -8.13 -8.33 0.525 -8.24 -8.35 0.380 -8.38 -8.52 0.292 -8.54 -8.68 0.195 -8.80 -8.86 0.104 -9.07 -9.09 0.000 -9.66 -9.58 aAll sodium-23 chemical shifts in this and following tables are referenced to infinitely dilute aqueous Na+. 1°0.1 M in salt. c . . : Isosolvation p01nt — 0.23 XTHF' d . . : Isosolvation p01nt - 0.15 XTHF' 49 Table 11. Sodium-23 Chemical Shifts vs. Composition of THF/MeOH Binary Mixtures at 25°C. 6Na+ XTHF NaBPh4(a) NaClO4(a'C) 1.000 -7.61 (i0.05) -8.39 (i0.05) 0.897 -6.58 -7.50 0.808 -5.99 -6.99 0.702 -5.57 -6.46 0.597 -5.19 -S.97 0.489 -4.85 -5.55 0.394 -4.59 -5.18 0.320 -4.47 -4.93 0.198 -3.46 -4.52 0.0997 (b) -4.19 0.000 (b) -3.87 a0.1 E in salt. bInsoluble. c . . : Isosolvation p01nt — 0.63 XTHF' 50 m To .3 I-x-I-e-l .mucHOQ COHum>HOmOmH A+VAtplquHumz fl H.o Axv uvzmmmz .mmusuxHE um\hm8 mo coHuHmanoo .m> umwnm HMUHEmso mmIEDHUOm HHS-I -— 090 11 1.411 11 V’l‘ ll th mKl le odwl IIIIIS IDOIwauo 0N9: .m wusmwm 51 s. H... .o. 1.68.. 58.630663 .leesemmz a e... .x. “608.62 . m .mmusuxfls momz\m:9 mo coHuHmOQEoo .m> unfizm HmoHEwso mmlsswpom m on: an ‘— di- 00.0 0.0! O.¢.I 0.01 0.0.: OK! 0.0) ms 10099890 9N);Z 52 in methanol). The isosolvation point values are listed in Tables 10 and 11. Differences in the isosolvation points of NaClO4 and NaBPh4 in PC-THF mixtures anadue to the interactions of the anion with sodium ion, i.e., ionic association. Isosolva- tion theory assumes no interaction between ions, only sol- vent-ion interactions. It was shown earlier in this chapter that NaClO4 strongly forms ion-pairs in THF while NaBPh4 does not. This difference in ionic interactions is re- 23Na chemical shifts of the flected by the difference in two salts in pure tetrahydrofuran. In neat prOpylene car- bonate the chemical shift of the sodium ion is independent of the anion. Previous workers have reported (67) that NaBPh4 does not form contact ion-pairs in propylene car- bonate. It should be noted, however, that the isosolvation curve for NaBPh4 does not display the usual behavior. In fact, the NaBPh data exhibit an S-shaped curve with vary- 4 ing solvent composition. Thus, there may be ionic in- fluences in the solvent mixtures containing this salt. Nonetheless, the isosolvation composition (low THF mole fraction) is in agreement with the higher Gutmann donor number of THF. In MeOH-THF mixtures, the data are more difficult to interpret. While NaClO4 is strongly associated in THF, the chemical shift of this salt parallels that of NaBPh4 as the composition is varied. This parallelism has been reported 53 by Popov and coworkers (61) in neat solvents. The only conclusions one may derive for this system is that of an approximate isosolvation point using the NaClO data while 4 recognizing that the anion has an influence. The value thus obtained is again in agreement with the higher donicity of MeOH. In conclusion, the isosolvation points have been deter- mined by sodium-23 nmr for PC-THF and for MeOH-THF mixtures. In all systems, the solvent with the higher Gutmann donor number had the lower composition at the isosolvation point. CHAPTER 4 KINETICS OF COMPLEXATION OF SODIUM ION WITH lB-Crown-6 IN NONAQUEOUS SOLVENTS 54 A. Introduction The observation by Lin and Popov (31) of an anion and solvent influence on the kinetics of complexation of sodium ion with 18C6 in tetrahydrofuran has remained unexplained. The authors observed slow exchange on the sodium-23 nmr timescale at room temperature for the system NaBPh4-18C6 in THF when the sodium ion concentration was greater than that of the crown ether. However, fast exchange was ob- served when ClO - or I- were the counterions. This is the 4 first known observation of a slow exchange at room tempera- ture for an alkali metal ion-crown ether system. In addi- tion, this is the first known observation of an anion in- fluence on the kinetics of complexation of sodium ion with a crown ether. The kinetic parameters for these systems were not determined in the above investigation. It is the purpose of the work presented in this chapter to investigate both the anion and solvent influences on the kinetics of complexation of sodium ion with 18C6. B. Choice of Solvents and Salts Shchori and coworkers (25) found an Arrhenius activa- tion energy of m12.5 kcal-mol-l for the decomplexation step of Na+-DB18C6 in MeOH, DMF, and DME solutions. They 55 56 therefore concluded that the major barrier to decomplexa- tion is the rearrangement of the complex prior to release of the ion. However, the three solvents in which their studies were done have approximately the same donor abili- 1t1es (DN MF=26.6, DN 25.7, DNDME=24). Therefore, it was D MeOH= important to examine the complexation kinetics in several solvents of varying donor abilities to test their hypo- thesis. Tetrahydrofuran was chosen to investigate the anion in- fluence on the complexation kinetics in this solvent since the low dielectric constant of this solvent obviously favors ion-ion interactions. In addition, THF has a donor number of 20.0 which is lower than those of the solvents studied by Shchori EE.21' (25). Methanol was selected to compare the complexation kinetics of Na+:18C6 with those of the crown ethers DC18C6 and DB18C6 studied by Shchori and co- workers (25) in this solvent. PrOpylene carbonate was chosen because this solvent has a donor number still lower than that of THF (DN = 15.0) but, on the other hand, has a high dielectric constant. We also studied the kinetics in MeOH-THF and PC-THF solvent mixtures to compare the results with those found in the neat solvents. Table 12 lists key solvent properties for the neat solvents studied. Sodium tetraphenylborate was the primary sodium salt used. As discussed above, this salt is interesting in that it exhibits slow exchange with 18C6 in THF at room tempera- ture. It was also our intent to minimize whenever possible, 57 oHng o.m6 ono\ xo .mm I mm.m o.ma um .66 I smm.o s.mm s.mm msoIO/ sows m .moal mmv.o o.om m.e on _ are .06. .Q.E .00. haemoomfl> umnEsz ucmumcou musuosuum ucm>Hom nocoo ccmEusw oauuooaoflo .Uomm um mucm>aom mEom mo mwfluummoum pmuomamm .NH magma 58 ionic association effects on the exchange process in all solvents studied. In THF and in PC this salt is known not to form contact ion pairs to any appreciable extent (SO-55,67). Therefore, it was used in all solvent systems with the exception of MeOH in which it is not sufficiently soluble. The choice of a sodium salt which exhibits fast ex- change with 18C6 at room temperature in THF was influenced by several factors. It is necessary that both the solvated and complexed sodium salts be sufficiently soluble to do the experiment. Unfortunately, complexed NaClO4 is not appreciably soluble (<0.05 M (68)). It should be noted, however, that the solubility of the complex increases if the salt concentration is greater than that of the crown (68). The complexes with NaI and NaSCN are soluble at reasonable concentrations (>0.05 M). The NaSCN was selected over the NaI since in the former case it is possible to in- vestigate ionic associations by infrared spectroscopy (see Chapter 3). In addition, this salt and its complex with 18C6 are soluble in MeOH. Since this was the salt used by Shchori EE.E$° (25) in their kinetic investigations, dif- ferences in kinetic results will be due solely to the type of crown ether used. 59 C. Results and Discussion The kinetics of complexation of sodium ion with 18C6 was studied in six neat or mixed solvents by using a complete 23Na lineshape analysis. The six solvent systems were THF, PC, MeOH, MeOH-THF mixture (40-60 mole %), and PC-THF (20-80 and 60-40 mole %) mixtures. The complete lineshape analysis requires information concerning chemical shifts and linewidths of the resonances corresponding to the solvated and complexed sodium sites in the absence of exchange. Therefore, this information was obtained and is presented here. In this discussion, site A refers to the solvated site while site B refers to the complexed site of the sodium ion. 1. Measurements in the Absence of Exchange The spin-spin relaxation times, T and the chemical 2A' shift, 6 of the solvated sodium sites in each solvent were AI obtained with solutions containing salt and solvent only. Since the sodium ion complex with 18C6 is very stable, i.e., 4 -1 Kf > 10 M (31,69), in all neat solvents used here and presumably the solvent mixtures as well, solutions in which the crown concentration is greater than the salt concentra- tion have only complexed sodium ions. Thus, these solu- tions were prepared to obtain T2 and 68' The temperature B dependence of T 0 and 6B are given in Tables 13-19. 2A' TZB' A’ .6666 :6 a 6.6 60 n O 666666+6z.\666 666666. 66666 6665 u .m.sm 66.66- .666 6.66 66.6- 6.66 6.66 66.66- .666 6.66 66.6- 6.66 6.66 66.66- .666 6.66 66.6- 6.66 6.66 66.66- .666 6.66 66.6- 6.66 6.66 66.66- .666 6.66 66.6- 6.66 6.66 .6.66. 66.66- .6666. .666 .66. 6.66 .6.66. 66.6- .6666. 6.66 .66. 6.66 .see.66 .6-6.66e\6 .60. 6 .see.66 .6-6.<6e\6 .60. e OHOH " AMVomoz o ” Anvomoz p.mcoflusaom mme :6 0066 £663 xmad I500 666 m0 0cm vnmmmz mo mmumm coflumxmamm 0cm mumwcm 66065650 mmIEsflpom .ma mHnMB 61 .6666 c6 mo.0 n 666666+6z.\666o6666666 66666 6662 u .6.26 66.66- .6666 6.66- 66.6- .666 6.66- 66.66- .6666 6.66- 66.6- .666 6.66- 66.66- .6666 6.66- 66.6- .666 6.66- 66.66- .6666 6.6- 66.6- .666 6.6- 66.66- .6666 6.6 66.6- .666 6.6 66.66- .666 6.66 66.6- .666 6.66 .6.66. 66.66- .6666. .666 .66. 6.66 .6.66. 66.6- .6666. .666 .66. 6.66 .26e.66 .6-6.66e\6 .60. 6 .see.66 .6-6.66e\6 .60. 6 “OOH vaomoz O " AMVomoz 6.666666666 666 :6 6666 s66: memEou mufl m0 026 zummz mo mmumm coflummemm 0cm muMHnm HMOHEOLU mmIEDHUOm .66 dance 62 .6666 66 m 6.6e .66666m+6z_\6666666o66. 66666 6665 u .m.sm 66.66- .6666 6.66- 66.66- .6666 6.6- 66.6- .6666 6.66- 66.66- .666 6.6 66.6- .666 6.6- 66.66- .666 6.66 66.6- .666 6.6 66.66- .666 6.66 66.6- .666 6.66 66.66- .666 6.66 66.6- .666 6.66 .6.66. 66.66- .6666. .666 .66. 6.66 .6.66. 66.6- .6666. .666 .66. 6.66 .sdd.6 .6-6.666\6 .60. 6 .sse.66 .6-6.666\6 .60. 6 e.6co66s6om 66 66 6666 6663 666d IEOU mufl mo van 6.3062 m0 mmumm COHummemm 0cm mumwnm HmoHEwnu mNIEDHUOw .mH manna .uHmm CH 2 H.0 63 Q .666666+6z.\6666666666L o6666 6662 u .m.s6 H0.NI N.00H N.¢NI Nv.VHI .0¢0N v.mHI 00-NI 0.HMH m.MHI 66.66- .6666 6.6- 66.6- 6.666 6.6- 0v.vHI .mmHH 0.0 0H.mI H.hm m.h 66.66- .666 6.66 66.6- 6.66 6.66 HH.vHI .HOB N.mm mv.MI m.vh m.mN AN.OHV NH.vHI AwOHHV .OBm AH“. m.mm AH.OH. vm.mI AmOHH. v.05 AH“. 0.Hm .sde.66 .6-6.666\6 .60. 6 .sde.66 .6-6.66e\6 .60. s @HoH H AMV m2 0 H A0vomoz 6.6662 :6 6666 s663 meQEOO mufl mo 0cm zummz mo mmumm coflummeom pcm mEMHnm HMUHEwSO mmIEs6pom .06 magma 64 .u6mm :6 m 6.on .6muoum+ng\6myoumoom6_ 066m“ m6oe u .m.zm No.mal .vhma m.ml vm.vl .Nha h.mI no.mHl .moaa ¢.ol No.ml .ooa m.o Ho.oHl .mvm v.w oa.ml .vva v.m av.mHI .mhh m.hH oa.ml .MMH m.hH mv.mHl .mmw «.mm wN.mI .hNH m.mm o¢.mHI .mHm m.mm mm.ml .HNH o.mm Am.ouv 0H.mHI Awoawv .mnv AHHV m.ov AN.OHV mv.m| Awoawv .VHH AHHV w.ov 65amvm© 66-mv mme\6 6006 a Asmmvam 66-mv «Na\6 600V 6 mo.6 n Amy.m.z o n Amy.m.z n.musux6z mmeumomz w m6os owuow m :6 oom6 £663 memEou mufl mo Gum vammmz mo mwumm coflumxwamm cam mamanm HMOHEan mmleswvom .ha magma 65 .6600 :6 m 6.0 n 0.06- .0006 0.06- 00.0- .006 0.06- 0.06- .0066 0.6- 00.0- .006 0.6- 0.06- .660 0.0 00.0- .606 0.0 0.06- .000 0.06 00.0- .066 0.06 0.06- .600 0.06 00.0- .666 0.06 0.06- .000 0.60 66.0- .066 0.60 66.060 0.06- 660660 .600 6660 0.00 66.060 66.0- 600660 .066 6660 6.00 6200000 66-mvm6ex6 6000 0 6000000 66-00060\6 6000 0 0 6.0606x6z 06-000 0 0602 06-00 0 :6 0006 0663 x060 IEOU mufi no can nmmmz mo mmumm coflumxmamm new muwflnm HmUHEmnu mmnesflvom .mH wanme .6600 06 m 6.06 .6muoum+mzu\63300606H 06006 m6oe u .m.20 66 mo.hl .va m.v~l HB.©HI .NmNN m.val mh.hl .mmm m.¢HI hm.mal .mHmH N.vl om.hl .th o.ml vo.hHI .wHHH N.m mo.ml .mNN m.m mm.wHI .mom m.va 0H.mu .mmH m.mH AN.OHV mh.mal AwOHHV .voo AHHV o.mN AN.OHV Hm.ml AwOHHV .th m.mN 6200000 06-00060\6 “000 0 0000000 66-00060\6 A000 0 mH.H H .m.2 o H .m.2 200 A63 0 0.0606x6z 000-06 0 m6OE 00-00 0 06 0006 0663 memEou mufl no new :mmmz mo mmumm coflummemm cam mamanm HMOflemzu mmIEswcom .ma magma 67 Figure 7 shows plots of 109(l/T2) vs. inverse temperature for NaBPh4, NaSCN, and their complexes with 18C6 in THF. Figure 8 shows such plots for NaBPh4 and its complex with 18C6 in PC and for NaSCN and its complex in MeOH. Figures 9 and 10 show the results obtained in the mixed solvents MeOH-THF (40-60 mole %), PC-THF (20-80 mole%), and PC-THF (60-40 mole %). Sodium-23 nucleus has a spin of 3/2 and thus, has a quadrupole moment. The dominant relaxation mechanism is through quadrupolar interaction modulated either by dif- fusion of solvent molecules in and out of the solvation sphere of the Na+ ion or by rotational diffusion of the complex. The quadrupolar relaxation rate may be written as (70): 2 2 -1_=_1_=_§_ 21” (1+§—)(39-——3V)2T T1 T2 4012(2I+l) 3 h 32'2 C where l/Tl is the spin-lattice relaxation rate, l/T2 is the spin—spin relaxation rate, I is the spin of the nucleus, 5 is the asymmetry parameter, Q is the quadrupole moment of the nucleus, 32V/BZ'2 is the 2 component of the electric field gradient at the nucleus, and TC is the correlation time which characterizes the fluctuations of the field gradient. For a simple reorientation process the variation of the correlation time with temperature may be expressed by (71): 68 . .c00 +mz cmxmameoo coma new cWWm>Hom mcwcfimucoo mcoflu usHOm cmusmonpxzmuumu now mmusumumasmu Hmooumfiomu .m> ~B\H no muon modasmm .0 06:00h 67000 7020:0006 00-16. 0...“. cm...” 05»... on...” and 07m. omd 6 _ . _ _ _ . _ . —i— —L— — ‘- _ 6 on... i. L. -:- 006 f OWN I. _.| O .- nu H\0\m\ 6 2. M\\\\l. .E-006 L \‘l ( \III .II 65 3 3 -:- otn _ ( -+- 000. - zom00m mzwz6mucoo ozoflusaom 60m mwusumummsmu Hmoou Moms .m> ~B\H no muoHQ moHHEom r-xov 702070006 4.- L: on...” om.n Oman _ _ _ _ 4|— - 00 z. 000. -. 0100,62 on. 2. .000“... 0:00.02 :00: z. 0006- 2002.2 10m} 2. :mmmu: ZUm epr-fi-HT m” 2 T2 298.15 where (l-) is the relaxation rate at 298.15°K, T is the 12298.15 absolute temperature, Er is the activation energy, and R is the gas constant. The values obtained for (%) and 2298.15 Er from such fits are reported, along with chemical shifts at 25°C, in Tables 20 and 21. The results for 0.2 M NaBPh4 solution in THF agree quite well with those of Ceraso (74) for a 0.4 M NaBPh4 solution 73 .6600 06 m 00.00 .6000 00 m 0.0 .0 mocmummmmo .6600 06 m 6.00 .6600 06 m 6.00 2000-000 000-00 06 000002 06.060 60.0- 00.6 .006 26.060 00.0- 00.6 .666 0000-060 000-00 06 000002 2000-000 06.060 06.0- 00.6 .066 000-0002 00 00002 26.000 06.0- 06.0 .000 00 06 000002 26.060 00.0- 66.6 6.00 0002 06 20002 06.060 60.6- 06.0 .066 A0000.6. 00 20002 00.0- 00.0 0.00 A300.6. 06 000002 06.060 00.0- 000600 060.0 200660 6.00 A0000.6. 06 000002 000000 2 600.6000060 006 60 00060600 HI AHImV AJWQ 0.00cm>aom Umuowamw :0 +02 pmu0>aom How muasmmm mmI530Uom .om magma 74 .6600 06 $.00.00 .6600 06 m 6.00 .6500 :0 $- 0.00 66.060 0.06- 66.0 .000 6000-000 000-00 06 0006.000002 66.060 0.06- 00.0 .600 6000-060 000-00 06 0006.000002 60.060 0.06- 06.0 .000 2000-000 000-0002 06 0006.000002 66.060 0.06- 0.0a .060 00 06 0006.000002 66.060 6.06- 00.0 .000 0002 00 0006.20002 00.060 0.66- 00.0 .000 0000 06 0006.20002 26.000 0.06- 600660 66.0 600660 .000 0000 06 0006000002 600000 6 600.6000000 006 60 00660600 HI 0.700 Tfl. M .mucm>aom Umuomawm £0 +02 Umxmadeou Mom muaSmmm mmIEDHUOm .Hm 00909 75 (see Table 20). The observation of a somewhat larger relax- ation rate by Ceraso may be simply due to the increased viscosity of the more concentrated solution. The difference in the results for THF solutions contain- ing NaSCN and NaBPh4 are quite striking. As discussed in Chapter 3, these differences are the result of ion pairing differences of the two salts. The relaxation rate of a quadrupolar nucleus is very sensitive to the symmetry of the environment around the nucleus. Contact ion pairing which occurs for NaSCN in THF (56) distorts the symmetry of the ion and thus increases the relaxation rate. Sodium tetraphenylborate does not form contact ion pairs to an ap- preciable extent (50-55) and therefore, a slower relaxation rate is observed. Ion pairing also influences the 23Na chemical shift (50). The effect of ionic association on the activation energy for solvent reorientation, although more difficult to interpret, must be recognized. The relaxation rate for site A in PC solutions is much larger than in THF, MeOH, or the MeOH-THF mixture. Other workers (67) have concluded that NaBPh does not form con- 4 tact ion pairs in PC. Therefore, this increase in the relaxation rate is most likely due to the much larger vis- cosity of PC as compared to MeOH and THF (Table 12). It is interesting to note that the value of Er for the MeOH-THF mixture falls exactly where one would expect if a linear relationship occurs between solvent composition 76 and solvent reorientation energy, i.e., Er . = mixture (0.6)(Er ) + (0.4)(Er = 1.37 kcal-mol-l. The inverse THF MeOH) relaxation time is larger in this mixture than in the pure solvents. This is likely due to the fact that both types of solvents molecules enter into the primary solvation shell of the ion thus distorting the symmetry around it. This distortion may also be causing the increase in the ob- served relaxation rates in the PC-THF mixtures but it is difficult to separate this effect from that of increased vis- cosity upon addition of PC. The complexation of the sodium ion by a planar macro- cyclic ligand, such as 18C6, also distorts the symmetry around the ion, thus increasing its relaxation rate. This effect can be seen in Table 21. Upon complexation of NaBPh4 by 18C6 in THF the inverse relaxation time increases by a factor of eight. The differences in relaxation rates, chem- ical shifts, and activation energies between complexed NaBPh4 and complexed NaSCN in THF solutions are due to the fact that complexed NaSCN is contact ion paired while complexed NaBPh4 forms crown separated ion pairs (Chapter 3). Of particular interest are the results observed in the solvent mixture as compared to those of the pure solvents. In the series neat THF---(60-40 mole % THF-MeOH)---neat MeOH the inverse relaxation time of the complexed ion in- creases while the activation energies go through a minimum. In the series neat THF---(80-20 mole % THF-PC)---(40-6O mole % 77 THF-PC)---neat PC the inverse relaxation times of the com- plex decrease while the activation energies also go through a minimum. The decrease in relaxation rates is unexpected considering viscosity influences only. No explanations for these observed trends are known at this time. 2. Fourier Transform NMR Exchange Equations The theoretical description of the effects of chemical exchange between sites A and B on the nmr lineshape has been given by Gutowsky 33 El' (75) and by Woessner (76). Ceraso and Dye (77) have modified the lineshape equations to in— clude some instrumental corrections. These equations have the form: G(w) = K[Icos(6O + 6') - Rsin(6O + 6')] + C SU + Tv _ UT - SV I ="§"“§ 3 R "—§___—§ s + T S + T P P A B T 5 0 __— + ——— + —————— - 1(0 + A - w)(wB + A - w) T2A' T23 TZATZB A P P A U=1+r(.—I.-§—+-.i.——) 2A 28 w +A-w w +A-w T = P + P 0 + A - w + T( A + B ) AwA B B T T 78 V = T(PBw + P w + A - w) M II 0‘13 T where 6(0) describes the lineshape as a function of frequency, 0, TA and TB are the mean lifetimes for sites A and B, 0A and w are the resonance frequencies for these sites in the ab- 8 sence of exchange, T and T are the relaxation times in 2A 2B the absence of exchange, PA = 1 - PB lation of site A, Gois.the zero order phase correction, is the relative pOpu- 6' is the first order phase correction, K is the intensity, C is the baseline height, and A is a frequency adjustment. The nonlinear least squares program KINFIT (49) was used to fit the nmr spectra to the equations of Ceraso and Dye (77). While the authors chose to fix the value of the first order phase correction, 9',we chose to adjust this parameter visually before the data transfer process, thus eliminating the need to include it in the fit. 79 Therefore, five parameters are required for the adjustment by KINFIT in order to fit the experimental data. These parameters are the intensity, K, baseline, C, zero order phase correction, 60, frequency shift, A, and the exchange time, T. Values for TZA' T23, 0A, and ”B were known from the measurements of the nonexchanging systems described previously and are entered as constants. Because the for- mation constants of the Na+-18C6 complex are large in all 4 1 systems studied (Kf > 10 M- (31,69)) all crown in solution was complexed with the cation. Therefore, and PB are PA known from solution preparation and are also entered as constants. Unfortunately, less than satisfactory results were ob- tained for the fits of experimental data. It was noticed that the severity of the problem increased with increasing experimental delay time. The delay time is the period of time which elapses between the application of a pulse to the nuclei and the collection of the FID. This influence on the quality of the fit may be seen in Figure 11. The problem arises from the fact that the relaxation rates of the two species undergoing exchange are very dif- ferent. In this particular example, NaBPh4-18C6 in THF at 25°C, the ratio of TZA/TZB : 8. Figure 12 shows the sum of two free induction decays with different relaxation rates. The first point of the FID determines the relative areas of the two resonances in the Fourier transformed 80 ........ hooohoooo‘ooookooooQooa 0.0.0 w- 3 ; -¢ = l : I I I Iro I I i I I . r I I g In I I .3 I I I i I l I 1 I' ' I I I I I I ‘ l. g I) I II ° I I I I I I I I I I II I II I “I I 0 I IIII II I I I I» II II 0‘ II : I II I III I I I I :- I“. u I II I II “a II .000 II I II IUOIOOII 08:: I II III Ia II II IIII I an. I III I I I I A t 0 L A ' = e I... T = : A I '0.-. .0..--...- o-.- 0.00' ---.'J ............. ,oaoo§--on‘uooo'-o..9.-”g ‘ ‘ = I I I I I a I f I t ' I I 6 I I I I I I f I I I I I I t l I. I I . I o I b . - I I 0 I I I I I IIII II no I 6000 II II OI I I II I :III I II I".- .‘ I 00 n I t I o .I II II IIII I I scgcoou one: I u I I II IIII .I I .II. I ‘ III I I I I oooobnoco9oooofi. p-00: . f : A = = - Figure 11. Computer fits of a sodium-23 nmr spectrum of a solution containing 0.2 M NaBPh4 and 0.1 M 1806 in THF at 25°C and a delay time of 800 us. (0) calculated point; (=) no difference between calculated and experi- (x) experimental point; mental points within plot accuracy. delay time correction: rection. (b) with delay time cor- 81 Intensity O-DE time ’ figure 12. ’Free induction decays with different relaxation times. 82 Spectrum. If no delay time is used, the ratio of areas in the transformed spectrum would be the one expected from solu- tion preparation. Experimentally, a delay time is required due to the fact that the applied pulse does not fall in- stantly to zero. As may be seen in Figure 12, during the delay time period the resonance which has a faster relaxa- tion rate decays a greater extent than the one with a slower relaxation rate. Since the first point of the FID is now at the end of the delay time, the apparent relative areas of the two resonances in the transformed spectrum are dif- ferent from those expected on the basis of solution prepara- tion. The relative area of the broader resonance is less than the predicted one due to the use of the delay time. The nmr lineshape equations modified to include chemical exchange, therefore, have been modified to include effects of the experimental delay time as well as the use of expon- ential linebroadening which is applied to increase the signal/noise ratio. The derivation is given in Appendix B. The modified equations have the form: G(w) = K{Icos[60-(wA+A-w)DE] - RsinEGO-(wA+A-w)DE]}+C I = AIMAG(XS) ; R = REAL(XS) c exp[(A -LB)DE] c exp[(A -LB)DE] xs _ 1 1 _ 2 2 ' A - LB A 1 2 ’ LB 83 C = 1(A2 + PAO'A + PBO'B) l C _ i(Al + PAO‘A + PBO‘B) 2 _ P P 2 4P P _ _ -l _ B _ A A B 8 AlpAz — [ (aA+aB+T i {(aA dB +-:F- -¥-) + 2 } ]/2 T _ ‘1 - OLA — T2A + 1(wA - w) = '1"1 + i( - ) 0‘13 213 “’13 m where DE is the delay time, AIMAG(XS) and REAL(XS) are the imaginary and real parts of X8, respectively, and all other symbols have their previously defined meanings. The appli- cation of these equations to the example illustrated in Figure 11a is shown in Figure 11b. As may be seen, the fit is much more satisfactory. Thus, these new equations have been used in all fits of kinetic data. It should be noted that the use of an experimental delay time in any nmr experiment will affect the relative areas of the resonances being observed regard- less of whether or not they undergo exchange. The cor- rection for the delay time effect in these instances is similar to that described in Appendix B and has been done for the Lorentzian lineshape case by Szczygiel (8). 84 3. Mechanisms of Exchange There are two mechanisms possible for the exchange of Na+ ion between solvated and complexed sites as proposed by Shchori gt §l° (10). These are the bimolecular process (I) and the dissociative process (II). 1 fia+ + Na+ol8C6‘——4> §a+olBC6 + Na+ I 8—"- k 2 Na+ + 18C6 ——A~ Na+ol8C6 II -2 The relaxation time for site i is given by .3; = rate of removal from site i TA number of molecules in site i Considering mechanisms I and II, then, _1_ 1 TA [Na+] (2 kl[Na+][Na+-18C6] + k2[Na+][18C6]) J; = 2k [Na+°18C6] + k [18C6] TA 1 2 Since K _ k2 = [Na+-18C6] f “'Ej‘ 2 [Na+][18C6] then 85 [Na+-18C6] [Na+] k2[18C6] = k_2 or +. £L-= 2kl[Na+-18C6] + k_2 [Na 18C6] TA [Na+] For the complexed site .3; = +1- (2kl[Na+-18C6][Na+] + k_2[Na+-18C6]) TB [Na ~18C6] l + Since i=i+i T TA TB then + l = 2k ([Na+] + [Na+-18C6]) + k (1 + LNa 'lscél) [Na J or + Na 3 1 _ + [ total ? _ ZkIENa Jtotal + k-Z + [Na Jfree The relative contributions of these two mechanisms to the exchange process may be determined at a given temperature 86 by plotting l/(IINa+]total) vs. 1/[Na+]free for several different relative free populations of Na+ ion. The slope will equal k _2, and the intercept will be 2k Assuming 1' either mechanism to be predominant a plot of log(1/1) vs. inverse absolute temperature will be an Arrhenius type plot since(lfi)ak, and the slope will be proportional to the Arrhenius activation energy. If the mechanism is known the rate constants may be determined. From the Eyring theory -AG# kBT As? — W) = T exp (72717) exp ‘ 7! AH ) RT where T is the temperature, AG# is the free energy of ac- tivation, AS? and AH? are the activation entropy and en- thalpy respectively, and all other symbols have their usual meanings. Also, in solution,AH#==Ea - RT. Therefore, all kinetic parameters may easily be determined once the mech- anism is known. 4. Results in THF Solutions The exchange time, T, was determined at various tem- peratures at two different free sodium ion populations for the systems NaBPh4-18C6 and for NaSCN'18C6 in THF solutions. Tables 22 and 23 list the T values for these systems and Figures 13 and 14 are plots of log(l/T) vs. inverse tem- perature for these systems. Figure 14 shows a plot of 87 Table 22. Mean Lifetimes as a Function of Temperature for the System NaBPh4-18C6 in THF Solutions.a PNa+ Temperature (°C) 1x103 (5) 0.412 26.0 ($1.) 7.10 (0.31)b " 36.7 3.72 (0.13) " 40.8 2.84 (0.08) " 45.8 2.02 (0.08) " 50.6 1.62 (0.07) " 57.3 1.06 (0.14) 0.735 25.2 (:1) 16.7 (2.1) " 30.2 9.82 (0.85) " 41.6 4.34 (0.27) " 49.8 2.59 (0.10) " 54.6 2.45 (0.21) " 59.6 1.66 (0.15) a0.2 g in s b alt. Standard deviation estimate. 88 Table 23. Mean Lifetimes as a Function of Temperature for the System NaSCN-18C6 in THF Solutions.a PNa+ Temperature (°K) 1x104 (5) 0.475 303.0 (:1) 0.8933(0.035)(b) " 297.8 1.048 (0.036) " 293.6 1.047 (0.041) " 288.0 1.363 (0.063) " 283.7 1.508 (0.077) " 278.6 1.493 (0.084) " 268.3 1.993 (0.140) " 267.1 2.203 (0.14) " 278.3 1.526 (0.10) " 274.1 1.748 (0.11) " 258.1 2.731 (0.38) " 264.0 2.707 (0.27) 0.735 309.4 0.9145(0.019) " 303.8 0.9878(0.020) " 298.6 1.050 (0.020) " 293.8 1.210 (0.031) " 288.2 1.401 (0.045) " 283.6 1.442 (0.05) a0.05 M in salt. bStandard deviation estimate. 89 .mco«u5H0m :muouomoxnmuumu cw moflumu maos mmomHH\me:mmmzu mooHum> um B\H .m> P\H mo muoHo moHMEom C(xov 765:8? and _ _ +4.26 mnnd +mo oumocmu .5om .IM 0 Humalzm Huaoe.amoxn .Ewficmnome mo cofiuofluommo How uxmu mmmo .Hlmm H .uo.o..H.HH ..H.H.a.m - .H.H. m.. .H.H. H.o .o...oon...o66H. . Hoo.o . H .vo.o.oc.HH 1o.H.m.o . .u.o.w°.a .o.o.uu.a .0...ooov..ooonm . mcq.c . H .Hc.o.cc.HH .a.~.a.p . .o.o..c.o .u.o.vo.o .o...oonnaaoonm . oHH.c omxo.o.mzsx..o HH .Ho.o.ov.HH ...H.6.HH- ...o.mn.m .c.ocma.a ...H..ona . 6.6.6 . HH .H°.o.c..HH .v.o.o.~H- .H.°.~a.a .H.o.H..cH ..HH..¢H¢ . HH..o . HH .Ho.c....nH .o.H.o.nHu .n.o.Hv.o .H.o.Ho.a ..m~..u~o . Ha~.o oax~.o.m=axo.o HH .Hc.o.Hu.~H .°.H.H.HH- .H.o.vH.a .n.o..a.a ..°a..°mmn . snv.o zoozx..c.maaxo.a H .Io.o.av.oH .H.H.H.HH- .m.o.sa.n .m.o.sn.. .o.¢ooop..oooonH ”cam mom.o on HH .Ho.c.HH.HH .°.H.H.h . .n.o.ac.a .H.o.ou.a ..cos..oomun . HH..° mom: H .Hcc.ocoo.cH .o.H.m.cH-. .m.c.oo.~ .n.o.m~.n .0...eo~H..o°nHm . nma.° . H .n.o.oe.cH .v.H.a.mH- ....o..o.~ .v.o.em.n .0...ooHv..oouma -zum ma..° . . HH .H.o. H.mH .°.H.H.HH- .a.o. ¢.HH .a.c. o.~H ..o..Hm .. mHH.o . HH .H.a. o.mH .~.H.H.HH- AH.o. H.HH .n.o. o.HH .H...HH..H6 .msmm ~H..° . was .6..somz .ncm66 .o.wm< ..n.m=< .num .awx coH:< +628 u=0>Hom .Uomm um coma nqu +mz mo coHumxmflmEoo mnu How mumqumHmm UHumcHx .vm magma 93 :0, (200- I?) (000- '6 ‘5 53-? 800- t! 3.. I. €“DC)P 2 400- 200- O0 4 8 (2 IS 20 ' / [Noflfree ‘Mfl’ Figure 15. Plot of l/(T[Na+]t ta ) gs. the inverse of the free sodium iog concentration for NaBPh4- 18C6 in tetrahydrofuran solutions at 25°C. 94 T(°K) _,‘ 20.0 298.0 LU) E' (8.0 E O| 5.9.,- I60 ° ' 285.0 2 “'3’ (4.0 - /i/‘}/ I\ V s 9 I20 - ° ______;, + 1 275.0 I0.0 - 8.0 1 l l l 1 0 IO 20 30 4O 50 '/ [Na+]free (M-I) Figure 16. Plots of l/(I[Na+]tota1) 35. the inverse of the free sodium ion concentration for NaSCN- 18C6 in tetrahydrofuran solutions at several temperatures. 95 .mumsflumm coflumw>mo Unmocmumm .mb mocmummwm IN m H- H.20 .EchmnomE mo cowumHHomwo How uxmu womb .5 mwo 02.. .mo . m H.. 0 HH mH.H ......................... Honv.n (Ho .H.o~= H .Ho.o.vo.HH .6.H. q.a - Ae.ocmm.o .q.o.nH.m .0...oon~..oonom . omxm.c.m=sxe.o HH .Ho.o.Hv.mH .q.o. e.~H- AH.o.so.m .H.o.w~.oH ..HH..mHo . umxH.c.m=sxa.o HH Amo.a.Hm.~H Ao.Hc H.HH- .H.c.q~.m .n.o.vm.m ..om..ommn . zomzxe.o.m=exe.o H .eo.o.mc.OH .H.H. ~.H~- .m.o.nm.n .m.oc~m.v 10...oooh..oooomH mama on HH .Ho.o.n~.HH .o.H.H~.H - .m.o.mo.m .H.o.mo.a A.OOH..oommm z :06: H AHo.o.mm.oH 1w.c. o.o~- .m.o.mn.~ .H.o.wm.n .0...oo~H..ooon (zom . HH AH.o. H.mH 1~.Ho m.~H- An.cc H.HH .n.o. m.HH .m...o..nm mnmm may . o 6 c a o 16.;662 .ncxuq lulxma .ncxza .6.» .mcx coHcc u=6>Hom .UomN um coma cswz +wz mo coflumonanu Ho“ muouOEmumm owuocflx mo umflq ommum>¢ omuzmwwz .mm mange 96 Chapter 3, and these kinetic results we are now able to explain the anion influence on the kinetics of complexation of sodium ion with 18C6 in THF solutions. In the bimolecu- lar exchange process two sodium ions must approach each other in the transition state. The contact ion pairing which exists for NaSCN and for its complex with 18C6 in THF is able to reduce the charge-charge repulsion which occurs between the Na+ ions in the transition state. Because in THF solutions NaBPh4 and its complex do not form contact ion pairs, the charge-charge repulsion is not offset and the dissociative mechanism is preferred. To be more precise, it is the degree of ionic associa- tion which determines which mechanism predominates. In THF solutions sodium iodide, perchlorate, and thicyanate all are strongly ion paired. They exhibit fast exchange and in all cases the exchange proceeds by the bimolecular exchange mechanism. Although this hypothesis was not tested for NaI and NaClO4, it follows that in the case of NaAlEt4 the dissociative mechanism should predominate since the degree of ion pairing of this salt is minimal (see Chapter 3). Although the salt used may contain some impurities, two solutions with different mole ratios of [Na+]total/[18C6Jtotal were prepared and the 23Na nmr spec- tra were recorded at 25°C. Since the total salt concentra- tion was not known the value for P was also adjusted in A the computer fit routine. The results are listed in Table 26. 97 .prm CH mmHuHHDQEH o: mcHEDmmw Nm¢.o n on .uamm :H mmeHHoQEH oc mcHEsmmm Hmm.o n +mzmm AHH.OHV om.m Amoo.oHVAnvmmm.o = = = = Awo.owv NN.H Awoo.OHvAmVNmH.o Am.ovm.mal Amoawvmam Am.ovma.h) Awoawv.mmm Ame OHxH mm Aeamvmm mme Aemmvmm an .UomN um mCOflgnaom mow: zommz mo coflpmxwamfioo on» How mumumamumm Ufluwcfix .mm manme 104 25, and 28 together with those of Shchori and coworkers (25) for comparison. It is interesting to note that the trend in the magni- tude of the formation constants for the 18C6 analogues in MeOH follow the trend in the dissociation rate constants, k_2. If k2 is diffusion controlled, the value of k_2 will determine the magnitude of the formation constant. In addition, the Arrhenius activation energies, Ba, and the enthalpies of activation, AHfz, for the dissociation step, follow this same trend. However, the entropies of activa- tion, Asfz, for the dissociative mechanism follow the re- verse trend and thus, the free energies of activation, Asz, are all essentially the same. One explanation for these trends may lie in the dif- ferences in the flexibility of the crown ethers. The flex— ibility of these crownsrmxnzlikely follows the order DBlBC6 < 18C6 f DC18C6 where DBlBC6 is the least flexible. Shchori and coworkers (25) have postulated that the main contribu- tion to the Arrhenius activation energy (and thus, Afifz) may be due to a change in the crown conformation. If this is true, it is reasonable to expect that the more flex- ible crown has a lower Ba and the more rigid crown has a larger Ea. This is the observed trend. In addition, if the transition states are similar, regardless of the crown ether, it is also reasonable to conclude that the more flexible crown will lose the most entrOpy in going to 105 the transition state. This is also the observed trend. The forward rate constants and free energies of activa- tion for DB18C6 and for DC18C6 are essentially the same. However, the forward rate constant for 18C6 is more than twice as large as for the other two crowns while the free energy of activation for the forward step is roughly No.5 kcal-mol-l lower. The greater flexibility of 18C6 over that of DB18C6 would explain this observation. The more flexible crown ether will be able to encapsulate the ion faster. The forward rate constant and AG? for DC18C6 do not follow the trends described above. The rate constant is lower and the free energy barrier is higher than those of the comparibly flexible crown 18C6. Assuming that the above assumptions are indeed correct, one possible explana- tion for this observation may lie in the structural dif- ferences of DC18C6 as compared to 18C6. The DC18C6 mole- cule (isomer B) has cyclohexyl rings above and below the plane of the crown cavity. These rings may result in a steric hindrance for the incoming Na+ ion, thus slowing the forward rate. Further studies are needed to elucidate these effects. 6. Results in PC Solutions Table 29 lists the T values, at various temperatures, for the system NaBPh4-18C6 in PC solutions. Figure 20 is 106 Table 29. Mean Lifetimes as a Function of Temperature for the System NaBPh4°18C6 in PC Solutions.a PNa+ Temperature (°C) TxlO5 (3) 0.505 30.5 (:1) 3.05 (0.37)‘b’ " 25.3 4.61 (0.40) " 20.8 4.36 (0.37) " 15.3 4.76 (0.38) " 10.9 5.61 (0.39) " 5.8 5.98 (0.36) " 0.5 7.94 (0.49) " -4.6 9.86 (0.70) " -9.1 10.1 (0.90) 0.341 -4.6 9.9 (1.1) 0.729 -4.7 7.31 (0.86) a0.1 g in salt. bStandard deviation estimate. 107 an Arrhenius plot for this system. The mechanism is de- termined from the plot shown in Figure 21. As may be deduced from Figure 21, the predominant mechanism of exchange is the bimolecular process. The kinetic results from the PC solutions are listed in Tables 24 and 25. It has been reported (67) that NaBPh4 does not form contact ion pairs in PC solutions. Thus, contact ion pair- ing does not offset the charge-charge repulsion of the sodium ions in the transition state in this system. Ap- parently, the high dielectric constant of PC (D = 65.0) is able to reduce this charge-charge repulsion and allow the bimolecular process to be favored. If this eXplanation is correct, then it remains to be explained why the exchange proceeds via the dissociative process in DMF (10,25) and in aqueous (78) solutions, both solvents having high dielectric constants of 36.7 and 78.5 respectively. 7. Comparison of Results in Neat Solvents Whenever the dissociative process is predominant, there is a large solvent influence on the kinetic parameters for the complexation of sodium ion with 18C6, as may be seen in Table 25. The free energy of activation, AGfZ, varies from 1 in THF to 7.2 kcal'mol-1 15.1 kcal‘mol- in aqueous solu- tions. Other kinetic parameters (Ea, AHfZ, Asz, and k_2) show equally impressive solvent dependence. However, in those systems in which the bimolecular 108 ll . mcoflusaom mumconumo mcoH>cobo ca mUmH.vnmmmz uOM B\H .m>.AHmuoum+mzHev\H mo uoHQ moHfiEom .om musmfim rsxov 705782 omfi. and 00.». and o _..n _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ 8n +cw may .m> Aamuoun+ngpv\a no uon .HN ousmflm ooLeFlA+oZv 0v on ON 0— o ll .oooom W l .l .2500? L_(ID’I°)(+DN)*nol) ll .oooonp .Ll .OOOOON 110 exchange mechanism is predominant, the free energies of activation are essentially solvent independent. In both cases the free energies of activation are ~10.7 kcal-mol'l. There are differences in their activation energies and en- thalpies. Schmidt and Popov (27) have also found that when the bimolecular mechanism predominates for K+ ion complexation with 18C6 in most solvents studied the free energies of activation are very similar. They found, how- ever, a large dependence of Ba on the solvent system. The activation entropy compensates this change in activa- tion energy to give essentially the same free energy of activation regardless of solvent. The results are ration- alized in terms of the solvation of K+ ion in the transi- tion state. Large AH: values are assumed to indicate weak solvation in the transition state. Since the solvation is weak, the ions cannot approach each other closely due to charge-charge repulsion. Therefore,lB: is more positive in weak solvating solvents since the crown has more room to "breath". In better solvating solvents, the ions may approach each other more closely in the transition state with the result of a lower activation energy and higher activation entropy. The same conclusions may indeed apply to our results. It is difficult, however, to say that the Na+ ion is more strongly solvated in the transition state of the system NaSCN-18C6 in THF solutions as com- pared to the PC results due to the strong ion pairing that occurs in THF solutions. 111 Cox and coworkers (79) observed a linear relationship between log(k_2) and the Gutmann donor number of the sol- vent for K+-cryptate dissociation. Sodium-23 chemical shifts were found to be even more correlated with the 39K chemical shifts (61,80). Therefore, donor number than a plot of loq(k_2) vs. solvent donor number was prepared for those systems in which the predominant mechanism of exchange is the dissociative one. Such a plot is shown in Figure 22. As may be seen the correlation is very good. Since log(k_2)°cAG7_‘2 according to the transition state theory, a plot of AGf2 vs. donor number was also expected to cor- relate well and is shown in Figure 23. The correlation between AGiZ and solvent donor number may be explained as follows: The free energy profile for the reactions may be ex- pressed as shown below: g AG §a+-Sn + Na+-18C6 —Q' §a+'sn + Na+-18C6 I + A652 Na '8“ + 18C6 _L II Na+-18C6 + nS where S = solvent In the dissociative mechanism, there is a net change in the number of solvent molecules in going from reactants to 112 .Amh.m~ .oa mmocoummmmv mUmHUQ A+Vlhw0mamo Ax. .mucm>qom mzoflum> ca :30uo.+mz How hopes: uococ consuou m> N x no uon moHfiEom mumEDz mozoo ZZ<§HDO o.mN _ _ . _ an 0.0 Cd o..v Dd o.m “coma .o. (toes) (Mam .NN musmflh 113 .Amw .mm .oH moocwumwmmv momaoo A+v qumHmo Axv “mmma on .mu:m> (HOm msoflum> ca csouo.+mz u0m umbssc Hococ camEusu .m> ~mo< no uon .mm ouzmflm mumEDz mOZOo 22.4.3130 Q? can . odu o9 — _ _ — , p . _ . _ i _ _ oo mow—8 a + II Qm oom.mo n x 82 .. 0 av II 0.9 .91.. Z 11 of LI odu 114 products. However, there is no net change in solvent mole- cules in the bimolecular exchange process. Thus, when the dissociative mechanism occurs, one must consider the solvent as an active participant. Certainly, the solvating ability of the solvent must influence A652 since in the transition state of the dissociative mechanism we have: A strongly solvating solvent (high donor number) will com- pete better with the crown ether and thus, reduce AGfZ. In contrast, one would not expect the solvent to have a large influence on the free energy barrier when the bi- molecular exchange mechanism is predominant. This is what has been observed by Schmidt and Popov (27) and by our- selves. Although the solvent does not appear to influence AG#, there is certainly a solvent dependence of AH: and of 08? as discussed above. Although one would not expect such good correlation between AG? and donor number in all solvents, the trend 2 does allow us to make reasonable conclusions concerning these systems. 115 According to Figure 23, if the dissociative mechanism were to occur in PC solutions, the free energy barrier would be m18 kcal-mol.l (DN = 15.). This would be the largest Asz value observed for a crown ether-alkali metal ion complex. The high dielectric constant tends to favor the bimolecular process, thus bypassing the need to proceed through such a high free energy barrier. Water has a much higher donor number (DN = 33.) and thus, a lower Asz for the dissociative process. Therefore, the predominant mechanism is the dissociative process in this solvent even though the high dielectric constant of water will support the bimolecular exchange mechanism. If we assume that ion pairing is negligible, predic- tions concerning the predominant mechanism in a given solvent may be made on the basis of the trends discussed above. The predominant mechanism will be the dissociative one in solvents which have both high donor numbers and high di- electric constants. The predominant mechanism will be the bimolecular process in solvents which have low donor numbers but high dielectric constants. In solutions of either low donor number and low dielectric or high donor number and low dielectric constant the mechanism will likely be determined by whether or not ion pairing occurs. Tetrahydrofuran solutions with either NaBPh4 or NaSCN salts exchanging with 18C6 are examples of this last case. 116 8. Results in MeOH-THF Mixture Table 30 lists the mean lifetime values as a function of temperature in a mixture containing 60-40 mole % THF- MeOH. The Arrhenius plot for this system is shown in Figure 24. As may be seen in Figure 25, the predominant exchange mechanism is the dissociative process. This was also the mechanism observed in both neat solvents of which the mixture is composed. All kinetic parameters for exchange in this mixture lie between those found in the neat solvents. However, the activation energy is closer to that found in neat MeOH solutions while the activation entropy is nearer to the value found in neat THF solutions. The composition of the MeOH-THF mixture is that which occurs at the isosolvation point (Chapter 3). Since the sodium-23 chemical shift has been found to correlate well with the Gutmann donor number of the solvent (61), it is reasonable to predict that the donor number for this mix- ture lies midway between the donor numbers of the neat sol- vents. Thus, the approximate donor number of this MeOH- THF mixture is (l/2)(DNT DN ) = 22.8. Figure 26 is HF + MeOH a plot of AGi2 vs. donor number, as shown in Figure 23, but with the result for the MeOH-THF mixture included. The cor- relation remains quite good. This result suggests two important points. First, it strengthens the belief that AGi2 is very dependent on the 117 Table 30. Mean Lifetimes as a Function of Temperature for the System NaBPh4'18C6 in a 60-40 Mole % THF-MeOH Mixture.a PNa+ T (°C) 1x104 (5) 0.477 31.8 (:1) 0.948(0.027)(b) " 25.3 1.24 (0.07) " 19.7 1.91 (0.04) " 14.0 2.42 (0.01) " 9.4 3.23 (0.09) " 4.0 4.84 (0.13) " -0.4 6.42 (0.095) " -5.9 9.15 (0.18) 0.744 25.3 2.61 (0.20) a0.1 M in salt. bStandard deviation estimate. 118 (sow 0 oz . v , I) .musuxas 8:6 1 z m H om1ov 6 :0 as H n fimomaea nmmmzH you axfi .65 pxfl mo soda soaasom .6N mucosa buxov 7622.89 8.». can on». open 9.». . T _ _ _ _ _ _ _ . _ _ _ _ _ _ _ CNN 1.. o; 1 O o - U V n _ (I own ( ) S 11. 3 O _ {Up 1). 8.... F. on». (TOU*(NO +)1010|)—1 vvuvo 60000. 40000. Figure 119 .1- n I! . l 1 I I l ‘ I T l 0 10 20 30 (N°+)- 1 free 25. Plot of 1/(I[Na+]total) 33° the inverse of the free sodium ion concentration for NaBPh4' 18C6 in a 40-60 mole % MeOH-THF mixture at 25°C. 120 ..mp .mm mucm>H0m msowum> aw :3ouo.+mz new amass: nococ ccmEusu mumEDZ mozoo ZZ<2HDO hpa mmocmummmuv m> mmwa no uon 06¢ Odn 0.0N odp 7)1 _ _ _ _ . _ fl _ . _ a , O O 23:: 017106: 5 89 u . mompoo I + .ITI DAM Oompmo n. x mowp I O .V .l) 06' nu . .3- Z ll.odp Ll.OdN .mm musmwm 121 solvating ability of a solvent as expressed by the Gutmann donor number. Secondly, and perhaps just as important, it strengthens the concept of isosolvation. We are able to calculate the approximate solvent donor number of the mix— ture based on the isosolvation curve. 9. Results in PC-THF Mixtures Tables 31 and 32 list mean lifetimes as a function of temperature for NaBPh4-18C6 in PC-THF mixtures (20-80 mole % PC-THF and 60-40 mole % PC-THF solutions). Figures 27 and 28 are Arrhenius plots for these systems. The predominant exchange mechanisms are determined from Figures 29 and 30. Results presented earlier in this chapter have shown that in THF solutions the primary exchange process is the dissociative one when NaBPh4 is the salt. In PC solutions, the predominant exchange mechanism is the bimolecular pro- cess. As may be deduced from Figures 29 and 30, the pre- dominant one is determined by the larger solvent component of the mixture. In the 20-80 mole % PC-THF mixture the pre- dominant exchange process is the dissociative one as is found in neat THF solutions. In the 60-40 mole % PC-THF mixture the predominant exchange mechanism is the bimolecular process as is found in neat PC solutions. It was concluded from the study of sodium ion complexa- tion kinetics with 18C6 in the neat solvents that high dielectric constant, low donor number solvents will tend 122 Table 31. Mean Lifetimes as a Function of Temperature for the System NaBPh4°18C6 in a 20-80 Mole % PC-THF Mixture.a p 1 T (°C) 1x104 (8) Na 0.283 40.8 (:1) 1.57 (0.04)(b’ " 33.1 1.95 (0.04) " 25.4 3.43 (0.05) " 18.9 4.41 (0.08) " 11.2 6.57 (0.14) " 4.0 10.1 (0.40) " -2.2 15.5 (0.7) " -9.5 23.4 (1.3) 0.415 40.8 1.82 (0.04) " 32.9 2.77 (0.03) " 25.3 4.64 (0.06) " 18.9 6.62 (0.10) " 11.2 10.1 (0.2) " 3.9 16.2 (3.8) " -2.2 24.8 (0.9) " -9.6 46.9 (2.0) 0.646 40.8 2.99 (0.04) " 33.0 5.01 (0.15) " 25.3 6.65 (0.16) " 18.9 9.03 (0.13) " 11.3 14.1 (0.3) " 4.0 24.0 (0.7) " -4.8 45.5 (2.0) " -12.5 99.5 (7.8) a0.1g in salt. bStandard deviation estimate. 123 Table 32. Mean Lifetimes as a Function of Temperature for the System NaBPh4-18C6 in a 60-40 Mole % PC-THF Mixture.a p + T (°C) 1x104 (3) Na 0.270 32.0 (:1) 0.666(0.032)(b’ " 24.8 1.07 (0.03) " 16.4 1.28 (0.05) " 9.5 1.87 (0.03) " 1.0 3.33 (0.07) " -8.5 6.43 (0.15) 0.406 32.4 0.723(0.04) " 25.2 1.05 (0.04) " 20.1 1.19 (0.03) " 13.0 1.65 (0.02) " 6.2 2.43 (0.03) " 0.2 3.57 (0.06) " -6.8 6.83 (0.18) " -12.2 11.7 (0.3) 0.663 32.0 O.996(0.02) " 24.7 1.17 (0.04) " 20.6 1.41 (0.04) " 14.6 1.57 (0.05) " 4.4 3.38 (0.04) " -5.9 8.60 (0.13) a0.1 M in salt. b Standard deviation estimate. 9.40 A I L) LIJ U) v 8.40 A I 2 (- v L9 0 ...l 7.40 6.40 Figure 124 .L u?” ‘3‘.\ " t . .__ ._\\g; " - a —_ 0 = 0.270 PNO+ -~ = 0.406 PM+ ‘ —)- I l l I I I I I I l I I I I I I I 1 3.10 3 30 3.50 3 70 3.90 1000475118-1 (OK-1) 27. Semilog plots of l/T vs 1/T for NaBPh4-18C6 in 60-40 mole % PC—THF sETutions. 940 8.40 A I L) LAJ U) 740 V A I :> ;S 6A0 V L9 0 _l 5.40 440 Figure 125 j_ 0.283 PNO+ .— X - 0.415 PNC+ j_ + -= 0.646 FIN-0+ T I I I I I I I I I: J I I I I I I I I I 3.00 3.20 3.40 3.60 3.80 100041949"1 (OK—1) 28. Semilog plots of 1/1 vs. 1/T for NaBPh4 18C6‘ in 20- 80 mole % PC- THF solutions. 126 om an soma.v:aamz .mm musmflm ooomN Um wwhsuxfla “$9-0m wIMHOp—ummml .m> AH um+mzuvv\a mo uon use :CLumuucmocoo :Ow EDMGOm mmum mmum>cfl may ootrlfl+ozv ON or .0 II .000— ll. .ooo¢ 127 .06mm 06 mmusuxas 009. I00 0 mace 0T 00 04 0003. 450062 now nodumuucmocoo cod ETHUOm ovum mmnm>ca ms» .m> AamuOum +mzupv\a no uon .om wusmflm 00.. LPIA+0ZV ov on 6N OP 0 _ _ _ . _ _ _ _ . . “I — a _ _ _ q _ J O I...- .OOOON II .0003 \I I1 0 n x...) II 00000 How (w .0... I 00000 max. A II .0000? 1 1 II .0000? 128 to favor the bimolecular process. As the PC composition of the mixture with THF is increased the dielectric constant most likely is increased. In addition, it is reasonable that the donor number of the mixture is decreasing upon increasing PC composition. Therefore, as observed in solu- tions with higher PC concentrations, the bimolecular process should predominate. It is interesting to note that in the 60-40 mole % PC-THF solution the free energy of activation for the bi- molecular process is essentially the same as those found in all other systems in which the bimolecular process pre- dominates, i.e., mll. kcal-mol-l. Thus, the free energy barrier for this mechanism appears to be independent of solvent. The Arrhenius activation energy in the 60-40 mole % PC-THF mixture is the largest observed for sodium ion com- plexation with 18C6 when the bimolecular exchange process predominates. The activation entropy for this system is the most positive observed for this mechanism thus far. It seems unusual that while exchange proceeds via the dissociative process in the 20-80 mole % PC-THF mixture, the free energy barrier is lower than that found in the neat THF. This was unexpected since the donor number of this mixture is likely to be lower than that of THF. Thus, according to the model presented above, the free energy barrier should be larger. However, the isosolvation curve 129 for PC-THF mixtures is not typical (Chapter 3). It is pos- sible that ion pairing or some other unexplained interac- tions occur at high THF compositions. Thus, this particular mixture may not be a good test of the model for the dis- sociative mechanism discussed above. CHAPTER 5 COMPLEXATION KINETICS OF CESIUM ION WITH DB21C7 and WITH DB24C8 130 Introduction The use of nmr techniques have greatly advanced the field of complexation kinetics. The exchange of metal ions be- tween solvated and complexed sites influences the nmr spec— trum of the ion nucleus. Through application of appropriate exchange equations one may extract the mean lifetime, T, of the ion from the nmr spectrum (see Chapter 4). Since T is related to the rate constant for the exchange process one may then calculate this value. Unfortunately, a problem arises in that many researchers tend to assume rather than demonstrate an exchange mechanism Since knowledge of the mechanism is obviously important in the overall study of complexation kinetics this practice should be discouraged. The usefulness of determining the mechanism of exchange is illustrated well by the results in Chapter 4. Had the mechanism been assumed to be the dissociative process in all systems investigated, the interpretation of the data would have been considerably in error. Recently, Shamsipur (30) investigated the complexation kinetics of cesium ion with several large crown ethers in acetone and in methanol solutions. The assumption was made that exchange proceeded via the dissociative process in all systems examined. Since both K+ ion (27) and Na+ 131 132 ion (Chapter 4) have been observed to exchange by way of the bimolecular mechanism, it is important to determine which process does indeed predominate for the large Cs+ cation. Such information may also provide clues as to how the cation size affects which exchange process occurs. Thus, it is the goal of the work presented in this chapter to discern which mechanism is predominant for cesium ion-crown ether complexation kinetics as investigated by Shamsipur (30). Specifically, the systems CsSCN°DB21C7 and CsSCN°DB24C8 have been studied in acetone and in meth- anol solutions. Results and Discussion A full cesium-133 lineshape analysis has been used to obtain all kinetic information presented in this chapter. Since a complete discussion of this technique has been presented in Chapter 4, only a discussion of the final results will follow. A. Measurements in the Absence of Exchange. Cesium-133 nmr chemical shifts and inverse spin-spin relaxation times for solutions containing solvated CsSCN and for solutions in which the ratio (crown)/(CsSCN)>l in acetone and in methanol, where the crown ether is either DB21C7 or DB24C8, are reported in Tables 33 and 34. Figures 31 and 32 are plots of log(l/T2) vs. inverse 133 Table 33. Cesium-133 Chemical Shifts and Relaxation Rates of CsSCN and of its Complexes with DBZlC7 and with DBZ4C8 in MeOH Solutions.a 0.02 M CsSCN T (°C) 1/T2 (s'l) 6(ppm) 25.5 ($1.) 7.5 (110%) -43.68 (£0.02) -50.4 6.3 -30.01 -60.1 8.5 -28.40 -70.3 11.0 -26.61 -80.0 10.7 -24.92 -88.9 15.4 -23.54 0.015 M CsSCN, 0.01644 DB24C8 24.7 7.5 -36.18 -29.8 22.9 -32.49 -40.5 29.5 -32.29 -50.6 38.9 -32.35 -60.3 43.7 -32.49 -70.4 51.2 -32.79 -82.7 87.0 -33.24 134 Table 33. Continued. 0.021 g CsSCN, 0.023 L4 082107 T (°C) 1/T2 (5'1) 6(ppm) 25.5 ($1.) 9.4 (110%) -l8.70 (£0.02) -l9.6 26.1 -12.98 -3l.2 33.3 -11.79 -44.3 78.5 -10.48 -55.9 69.7 - 9.33 -67.7 62.2 - 8.39 -80.2 92.4 - 7.58 aRef. to infinitely dilute aqueous Cs+. 135 Table 34. Cesium-133 Chemical Shifts and Relaxation Rates of CsSCN and of its Complexes with DB21C7 and with DBZ4C8 in AC solutions.a 0.02 I_I CsSCN T (°C) l/T2 (s-l) 6(ppm) 25.3 (11.) 8.5 (110%) -14.10 (10.02) -17.2 10.7 - 8.56 -28.2 8.8 - 6.94 -38.9 9.1 - 5.37 -49.8 10.4 - 3.73 -60.4 10.0 - 2.09 -72.4 10.7 - 0.16 0.02 M CsSCN, 0.022 M DBZ4C8 25.1 6.9 -26.62 -17.3 10.7 -25.53 —29.2 13.8 -25.16 -38.9 20.4 -24.89 -49.9 22.0 —24.59 -60.0 20.1 -24.23 -72.7 22.3 -23.74 0.02 1.4 CsSCN, 0.024 1.4 082107 25.1 7.8 - 5.93 -17.2 27.0 - 2.10 -28.1 53.7 - 1.15 -38.9 94.9 0.09 -49.5 94.2 1.49 -60.5 60.3 2.65 -72.9 54.0 3.74 aRef. to infinitely dilute aqueous Cs+. 136 I . I .Izom €339.48 m. No.0 A+v “Izum .movmmo.+mo 2 No.0 Ao. uzummo 2 No.0 hm. .mCOMuaaom Hocmzuwa :« :ofi Eswmmo pmxmfiaeoo 0:0 omum>aom How B\H .m> ~B\H mo muon moHfiEom .Hm musmam ATxov 76sz Tm «Am 5.? «xv 5n N...” “infill“..nilll 0.0 1' N.p .|0 O 5 . Dz .. .4 7w . . II It .. ( . ) .1 .0 ... Ir . II. x... /\ 137 .I200 .nwammo.+mo : ~0.0 1H0 “I200 .00¢~00.+mo m No.0 ADV “Aom mocmummmmv 020mm0 z no.0 Axv «20mmu 2 No.0 on .mcofluDHCn 020» I000 :M :0“ Eswmwo pwxmaasoo 0:0 pmum>aom 00m B\H .m> ~B\H mo muoHa moHflEmm .mm muswflm AFIxov TIazmh Wm 56 N6 5n N.n CF . 5.0 u - - -- - -I- — - .I -l 1 —— .1 - GI- +04 I-OI F04 (ks) ((30601 138 temperature for these systems. Since, in all cases, the 4 1 complex formation constants, Kf, are greater than 10 M7 (81), solutions in which the crown ether concentration is greater than that of CsSCN contain only complexed Cs+ ion. The inhomogeneity contribution to the relaxation rates is estimated to be :2 Hz. The 133 Cs nucleus has a spin of 7/2 and the predominant relaxation process has been shown to occur through quadru- polar relaxation (82). It is surprising to note, there- fore, that in several systems investigated here the inverse relaxation time does not vary exponentially as a function of the absolute temperature. In fact, in both AC and MeOH solutions the Cs+-DB21C7 complex exhibits a maximum in the plot. Unfortunately, Shamsipur (30) does not report his results for comparison. Deviations from linearity have been observed by 23Na nmr for solvated Na+ in pyridine (74) and in DMF (10). The authors suggested that changes in ion pairing might be responsible for the nonlinearity. Schmidt and Popov (27) observed deviations from linearity for solvated K+ ion, but not for its complex with 18C6, in 1,3-dioxalane, MeOH, AC, and an acetone-THF mixture. They reasoned that in the low dielectric constant solvents such as 1,3- dioxalane changes in ion pairing may be responsible for the nonlinearity. However, in methanol solutions, changes in solvent structure with temperature was suggested as 139 the cause for deviation from theory. In going from acetone to acetone-THF mixtures a reverse in the curvature was ob- served for which no explanation could be given. Mei (83) observed deviations from linearity of the 133Cs relaxation time for free and complexed forms of CsBPh4 in several solvent systems. However, the observed deviations may not, in fact, be real. The nonlinearity occurs at temperatures in which the relaxation rate is very slow and therefore, field inhomogeneity probably is res— ponsible for most of the relaxation process in this region. Her results for the solvated Cs+ ion in AC solutions are shown in Figure 32 for comparison. Our data are in agreement within experimental error if field inhomogeneity is con- sidered. In our case, the deviation from linearity of the relaxa- tion rate as a function of temperature is real. There is 133Cs chemical shift of a concentration dependence of the CsSCN in methanol and acetone solution at 25°C due to ion pairing (84). However, as shown in Figures 31 and 32, the effect of ion pairing on the relaxation rate of sol- vated CsSCN in these solvents at room temperature is minimal. Khaezali (85) studied ion pairing of CsSCN in methylamine at 25°C by cesium-133 nmr and found cesium linewidths always to be <2 Hz. The reason that the ion pairing does not significantly broaden the cesium-133 resonance is due to the very small quadrupole moment of 140 this nucleus (0.003 barns). As the temperature is lowered to the region where chemical exchange influences the cesium- 133 spectrum ion pairing effects are expected to become even more minimal due to the fact that the dielectric con- stants of the solvents increase with decreasing temperature (D = 29.5 at -50°C; D = 48.5 at -50°C (86)). It is reason- able to assume, therefore, that changes in ion pairing is not responsible for the observed behavior, especially since only the complexed Cs+ ion and not the solvated cation exé hibit nonlinearity. The lack of deviation in the relaxation times of the solvated CsSCN as a function of temperature suggests that solvent structural changes with temperature (which would affect correlation times) are not responsible for the ob- served nonlinearity of the complexed sites. The large in- homogeneity contribution to the relaxation rate of the sol- vated site makes it difficult to fully eliminate this pos- sibility. However, it is unlikely that both MeOH and AC would exhibit similar changes in structural prOperties as a function of temperature and in fact, the relaxation rates of the solvated sites in these two solvents do exhibit quite different behavior. Three other possibilities remain which may explain the observed behavior. First, the quadrupolar coupling con- stant of the complexed salt may be changing with temperature. However, since the quadrupole moment is so small for cesium 141 ion it is unlikely that this would have a large effect. Second, the correlation time of the complexed site may not be changing as expected. As discussed in Chapter 4, the correlation time, TC, is expected to vary exponentially as a function of temperature. Contrary to expectations, the correlation time will change if the structure of the complexed site changes drastically with temperature. The third pos- sibility is that there is an exchange process which is res- ponsible for the variation in the relaxation time of the complexed site. For example, an exchange between two dif- ferent confomers of the Cs+-DB21C7 complex may be slow enough to effect the 133Cs nmr spectrum. At low tempera- tures one confomer may predominate while at high tempera- tures the exchange is rapid between different forms of the complex. It is interesting to note that for complexed Cs+-DBZlC7 in AC and in MeOH solutions the relaxation rate goes through a maximum as a function of temperature, with the maximum relaxation rate occurring at essentially the same temperature and it is independent of solvent. Such behavior would be expected if the above exchange process were occurring. The Cs+-DBZ4C8 relaxation rate does not show such drastic deviation from expected behavior in AC or in MeOH solutions. Since the cavity of DB24C8 is larger than the diameter of the Cs+ ion, it is reason- able to expect that any conformational changes of the complex would be more rapid than that of the DB21C7 complex at any temperature. 142 Of the above possibilities to explain the nonlinearity of the complexed Cs+ ion relaxation rate as a function of temperature, the last two seem to be most likely. However, more study is needed to elucidate the cause of these in- teresting results. B. Kinetic Results Tables 35 and 36 list values as a function of tempera- ture for these systems. Figures 33 and 34 show plots of lOS(l/T) vs. inverse temperatures for CsSCN complexation with DBZlC7 and with DBZ4C8 in AC and in MeOH solutions at various relative free cesium ion populations. Figures 35-38 are plots used to determine whether the predominant exchange mechanism is the dissociative or the bimolecular process. In all systems investigated the bimolecular mechanism was found to predominate (the slope of the plots are essentially zero). The Arrhenius activation energies and other kinetic parameters were calculated from the data and are presented in Table 37, along with those of Shamsipur (calculated from his data on the basis of the bimolecular mechanism) for com- parison. As may be seen, all results are in agreement with those of Shamsipur with the exception of the system CsSCN- DB24C8 in methanol solutions. The origin of this discrep- ancy is unknown, but may be due to the small difference in chemical shift between solvated and complexed sites for this system. With the approximate nmr technique used by 143 Table 35. Mean Lifetimes as a Function of Temperature for CsSCN-Crown in MeOH Solutions.a Crown PCs+ T (°C) 1x104 (3) 082408 0.378 -51.0 (:1) 7.80 (0.42)(b) u " -59.4 7.75 (0.31) n u -57,2 11.2 (0.2) u " -76.4 16.0 (0.3) u n -80.8 24.4 (0.6) " 0.684 -59.6 7.99 (0.44) n n -67.2 11.4 (0.7) DB21C7 0.383 -19.9 1.62 (0.04) n " -31.2 3.64 (0.07) n . -36,6 4.83 (0.19) n " -45.4 7.95 (0.20) n " -60.7 20.7 (0.9) " 0.686 -19.8 1.69 (0.05) n n -35.7 5.65 (0.15) n " -43.2 8.02 (0.24) n n -55.1 16.9 (0.5) u " -6S.6 27.1 (1.3) a0.02 M in salt. bStandard deviation estimate. 144 Table 36. Mean Lifetimes as a Function of Temperature for CsSCN-Crown in AC Solutions.a Crown PCs+ T (°C) TxlO4 (s) 082408 0.384 -17.8 (:1) 0.417 (0-013)(b) u n -34.2 1.11 (0.04) n n -50.0 3.51 (0.08) n n —60.4 8.84 (0.17) n n -72.5 21.3 (1.0) " 0.608 -l7.7 0.536 (0.012) n n -28.8 1.09 (0.06) n n -39.2 1.81 (0.04) n n -55.4 7.05 (0.22) n n —72.6 24.8 (1.6) DBZlC? 0.385 -17.7 1.49 (0.04) n n -29.6 5.81 (0.26) n n -39.0 9.61 (0.22) n " -49.9 30.0 (1.0) u n -60.5 76.3 (4.4) " 0.663 —l7.7 3.34 (0.10) n n —28.5 5.70 (0.20) n n —39.1 14.4 (0.6) u n —49.9 28.9 (0.7) n n -60.4 66.2 (3.2) n n -72.5 261. (41.) 30.02 M in salt. bStandard deviation estimate. 3.60 ,1 l 8 U, 320 v A I 2 ~ 80 +- A. \J L9 0 __I 2.40 2.00 Figure 145 _- E: p I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I a I I 3.80 4.00 4.20 4.40 4.60 4.80 5.00 5.20 5.40 5.60 1000.I.7-'5.\/.F>-1 (024‘ 1) 33. Semilog plots of l/T vs. 1/T for CsSCN°DBZlC7 and for CsSCN-DBZ4C8 13 methanol solutions. (0) [CsSCNJ/[0821c7] = 3.18; (x) [CsSCNJ/[DBZlC7] = 1.62: (o) [CsSCNJ/[082408] = 1.61; (+)[CsSCN]/ [DBZ4C8] = 3.16; all DBZ4C8 data used for fit. 146 4.4.0 i}- 4.00 T 1. A 3... —L I O .. Lu (1) 3.20 -—- v A .. 'l' 280 — 2 -I)- ‘6 2.40 -- O A unl— 2.00 ‘1- 1.66—TIT 1.20 I 3.80 Figure 34. 1000*TEMP-1 (OK-1) Semilog plots of 1/1 gs. l/T for CsSCNoDBZlC7 and for CsSCN-DBZ4C8 in acetone solutions. (o)[CsSCN]/[DBZ4C8] = 1.62; (+)[CsSCNJ/[DB24C8] = 2.55; (x)[CsSCN]/[D821C7] = 2.97; (o)[CsSCN]/ [082107] = 1.62. 147 .XOOON Axv “xomaw .ov .mmusumumdemu msoflum> um.mcoHusHmm mcoumom ca nuammo.zummu qu coflumuucmucoo co“ Esfimmo mmum wmum>cfi 0:» .m> AamuoDm+mugev\H mo muon .mm musmwm 000 LPIA+mov CON om— cop on o — . _ . —I — n p .r — — . m P — — — u u u .o _ _ .. _ .F _ II .0000— -I m. n \4/ 0 S J! 0008 NH .9 m. II _ II 00000 II 600084 148 .XOOON Ax. “x 0mm A0. .mwuaumumdsmu m50w00> um mcoflusamm Hocmmume ca FUHNmo.zommu u0m :owumuucmocoo cofl Enammo mmum wmum>cw ms» .m> Adm» um+moupv\a mo manm .mm wusmwm 00.. 0PIA+moV 00m 00. 03 00 0 . . . . . . _ n “ .r u . _ u H . . u w u _ n u . u w a. O 3- m W II .0009 L- x) 1. 0 AI .0000“ w ) O S I- + ( 1. m. 4| 00000 NC I. _ III Jl. .OOOOV Ir II .0000... 149 .xooom .ov uxomam Ax. .mmuzumumdemu msoflum> um mCOAHSHOm chuwom cw muvmmo.zummu 00m :ofluouucoocou cow sawmmo mwum mmum>cw may .MN.AamuOum+mUuev\Hmflumuon .00 ouzmflm 00.. cPIA+moV 00m 09 00, on 0 “11111111111111"111.0 3- I) a II .0000N L- ) \IOI 0 HI 0000.. \m/ 3 S I + NH O ‘Iol JI 00000 max ._. 3- 4| 00000 1.. II 00000. 150 XowON Axv .xovam A0. .mwusumummEmu macaum> um macausHOm Hocmnuwe :a movmma. .zummu uOu :oaumuacmocoo cod Enammo mmum mmum>ca msu .m> Aamuoum +mUHpV\H mo muon .mm musmwm 02 0 IA+m00 com on? 09 on o — . _ _ _ — p _ . — — p _ p p — p p . _ — u _ . . — u q u - — - q _ u — a u q - D II .oooom flu 1.. D n h/ 3 S H II .0003 {W m. L- ... a % 1.. .808 H [I .0008 151 .mumsflumw :Oqufl>mv cumccmu .om mocmnmmmmm I0 Humauzo 5 mg HIHOE.Hmuxm A0.m0 0.m A~.000.0 Am.~0~.mmu Am.000.m Am.00m.m = = A0.m0 m.0H “0.000.00 A0.~00.0~- A0.000.m A0.0Vm.0 A2000.80 = A0.00 0.m AH.0vm.0 Am.avm.0au Am.000.0 Am.000.0 = = A0.00 0.H AH.00m.0 Am.~00.0fiu Am.0vfl.0 Am.00~.0 Amvnoamma mom: A0.H0 0.0 AH.000.0 Am.H00.mu Am.000.0 Am.0vm.h = = A0.H0 m.m AH.000.0 A0.~00.H- A0.000.0 A0.00N.0 Amvmuemmo = Am.0000.0 A~.000.0 Am.~00.mu Am.0va.m Am.000.0 = g Am.0v 0.H AH.000.0 A0.m00.0 A0.000.0 0A0.0vm.0 Amvsuflmmo u< 0Hxax 00 m0 00 mm czouu .>Hom onvl Amvx Anvx Amvx Amv .VmOONN Um WCOHuHaHOmw mom: Cum USN Ufi cw mumcum csonu Hmum>mm suflz +mU mo coflummeQEoo map How mumumEmumm oaumCHx .hm magma 152 Shamsipur (30), positive errors in the activation energies have been shown to occur when the chemical shift difference of the exchanging sites is small (87). It is interesting to note that the free energy of ac- tivation, AG#, is independent of solvent for the two crown complexes. This is due to lack of active solvent participa- tion in the exchange process (see Chapter 4). However, there are solvent influences on the activation energies and entropies of the kinetic process. The activation energies are higher for the two complexes in acetone solutions while the activation entropies are more positive in this solvent. Since acetone has both a lower donor number (DN = 17.0) and dielectric constant than does methanol (DN = 25.7), it is reasonable to assume that in the transition state acetone cannot reduce the charge-charge repulsion of the cesium ions as well as methanol and thus, the activation energy is higher in this solvent. The activation entropy is more positive in acetone due to more breathing room of the crown in the transition state due to the larger separation of the Cs+ ions. This observation has been made by Schmidt and P0pov (27) for K+ ion complexation kinetics and by us for sodium ion complexation kinetics with 18C6 (see Chapter 4). The larger crown ether has the larger rate constant and lower AG#, compared with DBZlC? regardless of solvent. Such behavior is expected since cesium ion has approximately the correct size for the cavity of DB21C7 but is smaller than 153 the cavity of DBZ4C8. The larger formation constant of the Cs+-DB21C7 complexed vs. that of the Cs+-DB24C8 complex 25°C _ AC,DBZlC7 ‘ = 3.96; also reflects the cavity size effect (logk 25°C 25°C AC,DB24C8 MeOH,DBZlC7 = 3.60 (81)). 3.98; logk 25°C MeOH,DB24C8 It is known that in these solvents the solvated CsSCN is = 3.71; logk logk ion paired to some extent (84). Unfortunately, information is not available concerning the degree in which the complexes are ion paired. Thus, it is not possible to determine whether ion pairing allows the bimolecular exchange process to predominate. If we compare the complexation kinetics of the alkali metal ions with crown ethers a trend emerges. As one goes through the series Na+--K+--Cs+ the predominant exchange mechanism varies from that of either the dissociative or bimolecular process for Na+ ion to primarily the bimolecular process for Cs+ ion. Assuming ion pairing is not responsible, this is likely due to the decrease in charge density as one goes to the larger cation in this series. Such a decrease in charge density will minimize the charge-charge repulsion in the bimolecular exchange process and thus, allow it to predominate. It would be interesting to investigate Li+ ion complexation kinetics to see if the above hypothesis is correct. One would expect that this ion would exhibit primarily the dissociative mechanism in the absence of ion pairing. 154 In summary, it is important to investigate not only rates of exchange involved in the complexation kinetics crown ethers, but also the mechanism of exchange. Such formation is invaluable in determining the kinetic pro- cesses of these complexes. the of in— APPENDICES APPENDIX A DATA TRANSFER PROGRAMS FOR THE WH-180 NMR 155 The programs MOVE and NTCDTL were used to transfer data out serial port B of the Nicolet 1180 computer on the WH-180 nmr spectrometer. A listing of MOVE and an example of its use are shown below: 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 350 360 370 380 390 400 410 420 DIM A(2047),X(8000),R$(6) PRINT "FIRST FIVE CHARS OF FILENAME"; INPUT R$(0) PRINT "INPUT FINAL FIVE CHARS OF FILENAME": INPUT R$(l) LET R$(2)="\" LET N2=8000 CALL BDEFINE(10,R$) CALL FREAD(10,X,N2) CALL FREAD(10,X,N2) LET N1=N2 CALL PAINT (Nl,l) CALL IAFLT(x,N1) PRINT "INPUT # 0F POINTS"; INPUT N PRINT "STARTING POINT"; INPUT 3 PRINT "ENTER STEP SIZE" INPUT 31 LET w=N+s FOR K=S TO w LET L=K-S LET A(L)=X(K) NEXT K CALL FDISP(A,N,800) PRINT "OK"; INPUT B$ IF B$="N" THEN 230 FOR I=S To w STEP 81 PRINT #8:I,X(I) NEXT I PRINT "DONE" END 156 157 Example of Move RUN MOVE FIRST FIVE CHARS OF FILENAME?NB251 INPUT FINAL FIVE CHARS OF FILENAME?.S INPUT # OF POINTS?600 STARTING POINT?2500 ENTER STEP SIZE ?2 OK?Y DONE The spectrum to be transferred is stored along with the language BASIC on the hard storage disk. Program MOVE is run and the appropriate input given. In the example, the data to be transferred are in the file NBZSl.5. The number of points in the region to be transferred is 600. The starting point of the region is 2500. The step size, 2, tells the program to skip every other point. At this time, the program will display the region to be transferred determined from the input above. If the region is satis- factory, a "Y" response will begin the data transfer. NTCDTL is a program written by Nicolet to transfer data out serial port B. This program was used when the WH-lBO software was upgraded in 1984. The following example demon- strates its use: 158 NTCDTL Example RUN NTCDTL DATE-TRANSFER PROGRAM VERSION #10903 COMMAND: BI,BO,CP,AP,AR,KB,LP,LR,MO,TL,TT? BO WHAT FORMAT (A=ASCII,B=BINARY)? A WHAT PARITY (E,O,M,N)? N MAXIMUM RECORD LENGTH = 64 PROMPT = 12 ENTER TERMINAL MODE (Y,N)? N - Hold down "Line Feed" Button on the floppy drive - The region of the spectrum to be transferred is chosen using the upgraded software. The program is then run. The command "BO" refers to serial B out. The data are trans- ferred in ASCII format. The parity chosen is none (N). The prompt "12" requires a line feed character prompt from the PDP-ll computer for each point transferred. Once the "N" response is given for the terminal mode question, the first data point is transferred. The "line feed" key is held down on the PDP-ll to transfer the rest of the file. APPENDIX B FT-NMR TWO-SITE EXCHANGE EQUATIONS MODIFIED TO INCLUDE LINE BROADENING AND DELAY TIME 159 Using the formalism of Gupta 23 31. (75) the magnetiza- tion may be written as: Azt A t 1 2e G = G + G = C e A B 1 + C where G is the total magnetization, GA and GB are the magnetizations of sites A and B respectively, t is the time after application of the pulse, and all other symbols have the same meaning as in Reference 75. The Fourier transform, S, of the free induction decay (FID) is given by -i(w-wrf)tdt S = fb Ge To modify the equation for the loss of signal during the delay time period (the time between application of the pulse and collection of the FID the integration is taken from t = DE to t = w. Thus w -i(w-wrf)t S = 1’ Ge dt where DE = delay time. When line broadening is applied to the FID, the FID is multiplied by exp(-LB-t) where LB is the line broadening. Thus, 160 161 = m *i(w-wrf)t -(LB)t SLB Q)Ge e dt = fmGeE-LB-i(w-wrf)]tdt 0 Including both line broadening and delay time S = f” GeE-LB-1(w-wrf)]tdt DE,LB DE Integrating gives the spectrum in the frequency domain: _C GIAl-LB-i(w-wrf)]DE S = l - DE,LB Al-LB-i(w-wrf) Az-LB-i(w-wrf) C2e[A2-LB-i(w-wrf)]DE where all symbols are described in Reference 75. Redefin- ing “A and dB as a = T.1 + i(w —w) a = T-1 + i(w -w) A 2A A ' B 2B B Then, _Cle(A1-LB)DE _C2e(A2-LB)DE S = - DE,LB Also, using the relationship 162 T T A B T = = T P - T P TA+TB A B B A Then, _ _ -1 -1_ -1 2 + 412P’1P‘1]%}/2 A B Using these new definitions and S we can program the DE,LB computer in complex Fortran to solve the spectrum for T. Szczygiel has shown (8) that the delay time modification introduces a first order phase influence on the transformed spectrum. This has the form 01 = cos (m-wC)DE where w is a constant. Due to instrumental contributions C to the first order phasing of the spectra which could not be included in the above equations, the first order phasing was done visually. The modified exchange equations used had the delay time contribution to the phasing eliminated when the spectra were fit. The nonlinear least squares program KINFIT (49) was used to fit the data to the above modified equations. The sub- routine EQN for the program is listed on the following pages. . _ 163 CCCCCCCCCCCCCCCCCCCCCCCCCCC CC CC . CC CC . 8888888 0888808888 AAQACC CC 888 8 NNNNNNN NNNNNNNNNN NNNNCC CC . NNN N 0000000 0500000000 0000CC CC 000 0 9C ) ) o-C CC I o 0 n.- CC CC A0 0 U 5 CC CC N88 3 08 CC CC 8 OP 8)! CC CC 6 8.8.0 000 «CC ) CC A o O 92L F18 CC N CC I. A ) U I O 0 88 9 CC 0 CC A N ) 0P8 [)8) CC 8 CC L A O 8 v- OP R0 00 CC 8 CC I L ) 8 OUU F0)5 CC C CC . X )) 8 KC" .308 CC C CC )) 8 8 8) 5 N 0 0 L808 CC 0.) ) CC )) I I 85 N) 008 A031 oCC .aN ) 5 CC 88 ) ) 88 02 UFA 978! CC 0.0 A N CC 88 U C 58 C8 N 00 X 080 CC C8 8 CC UU H H N5 0 U OPJ [)5 0 CC 8 C N CC 9 O P P UN )0 PO 0 LOX) CC CC 8 )C CC 88 2 L L C0 )) AF8 AU 90 CC 5.... la C0 CC 9.9. I A A o C 0) V 0A 93).: CC AR) 5 NA CC 0 O o O 0 A . 8) 080 C808 CC NRC 80 CC )) ) ) ) NU U8 NIN CP2P CC P08 :- 80 CC 11 ) A 8 AN 9 8 0X 0 U088) CC CA 0 5 CC 88 5 8 8 LA 8.8 808 RFL85 CC ) N N 8 CC xx 8 8 8 Val. P5 P .5 3 0A!) CC 8)CV 0 AC CC XI U 5 5 8x 9 N 0PN 8)VN8 CC 8COCC. P LN CC . u I N N ’8 )0 NYC 505 0X CC N INRNG 0 C... CC )) ) 0 0 )1 1C 08C N88)X CC 0 SIUCN P 08. CC 35 A C C )) 8 0 RX 0 OXCUXP CC 8 o NL UA 8 88 CC 88 8 O o )) X) CR8 2! 05 08 CC 8C CCOOH CC 88 8 A A 8) I) N 85 C )8)8 CC C" 8SNCC CC 55 s H H 88 c N IXC 0)0N68 CC C) NAP-RX CC NN N) P P 88 )8 RXN 1028.... CC R8 8BZFC CC 00 O) L L 58 30 0 O I C08N8 CC R 88888 CC CC C5 A A N5 8. PP8 93058) CC 08 A 8 CC 88 . 8 o 0 ON 88 AYC R. .8852 CC CA ACABZUCB CC 0 c )U ) ) C0 5P L8C NA 0 ON... CC L AA P8UP8UUL CC )) 50 1 1 a C N o Olav A8))0 8 CC "C 88 CC 0 o 8) 8 8 ) O 0) 8 0 v LXOOC) CC 80 CLU .. 8 8 8 8 8 a : CC 11)U5 8 8 )) C1 058 .5805 91 CC 8 "CA CC 0 0)]8 5 5 8) 88 8P0 I 0038) I CC U0 KC8U8 )))))))) CC 0 .5)U N N 88 . X of. 9 «AA .8800 CC N 82305680 CC 00888 0 0 88 )X E 0v. XH)L551 CC EA .. .. = = : 88888888 CCO 88U8/ C C 58 3 . 5 PR 0 0A02C8 0 CC 5 88888888 CCU Ina/8) o 9 N5 8)... AAJ 8L2 GNU-9 CC NC ))))) 55555555 CCU LL 05‘ ) a A UN 03 o 88J P88) 08 0 CC AN 12385 NNNNNNNN CC... PP1N8 o N N C0 880 8 o J OJ 0 00))80 .CC H.) 88888 00000000 CC "I. 90.8 2 A A o C N88 R 8 00 0 J3 926 0 0 CC CN UUUUU CCCCCCCC CCU CCBCS I o 0L L A . 85 A 8 C8" H OH) 01).! CC NC . CC8 O O H § N )22“ x NB 5N0 N F P5 9 889.00 00 9 CC CU CC ))PBU val/8 8 AN 0 00 I. N AEL EPLUZUSA CC A . CCU 25LHC 7.)). a LA PC ) U 8 8R 0 N0A3858 0 CC v.0 CCC 88APO LLL) ) NC A8) A‘ o R 0K N8 08 8K 0 98885 CC 8R CC) 88OL) AAA 0 0 8X )C . 1 8 R 0 CO: 0 9V0 IIAACSP 0 CC 85 5 CC 0 55AA138 O c 1 A 88)5R) - N [C UGO 58A 0 88H8C808 CC 5 R 5 CCU 9NNH . 8057:... O O P85X . 3 o N 5L AL8 N 08 DNPXVSL 0 CC 0 C C 8 . CC 0 500PA8BUCC o o XPX888C . 51 E F EU... OSEIL 0N 03 CC ON 8 . N CC81 ICCLHSQ LUUC 0 5X85. UN C AF R 9" N0 088ROAN)0) 0 CC .18 C . A CCCU QIIAPNJCRR8 8 C...LA)8 o L OR 9 o 981. IKTSSFP UICOZCCI C C CCC 8L N 8 CC 00C) 0 08L0r.8aBX I. 8CAH15H CA R822R8K o 8 SIN/[882 058U65 U52 U UCC. A 5 CC)0U 01.1. . ACRD 88L L o 8E8808 U: ASGGEF A 0 UN... QONNES 0)8 N: :NN : :NNNNCC RR R N CC5 0N3 .. : : 80FX : : P)P)8.t RAUCCNN:N C OCEUN 0 08R 008LCOOLNUO808EERIKRR8R8CC H0 A 0 CC8:80 ABC:8 :CABHBHUC) : = : a NRIUR RPPIICOF... RM. 08 OHWP...21P88DPJ8NA.J8.J8. .C 8.? P C CCU)82H..1CU:8. .Nu...l..CI.. . ZPCC....IU88.J UNXVv-HNHN88U NAA8NUV8N8NCC UU08G¥0008PYHUO88EOOOEOEOCC RSCUIYCCCD 00 CCIJRCNNRCRCCC 0CAACK8CNO DDRN88N "LIN RRON 9 Q 9 80 : KC AAUAHDHNL“. R A 8.23 123 CC CC CC UCNCR O I 0 C 1 7 a 2CC CC 5 CAI-AU 900) o . a-C CC 0 030) .8 .188 I-NAAJ .’ 0 o. u C. a non» NUAA8...8...A8005 5 CC CCCCCCCCCCC CCCCCCCCCCCCCCCC PJPHARARHFLLL? 1030 111 1 CC85N : PPUCCC 3 0A 2 A = 0 .- CAALR88NS8 CCF808LLRROLLLL1L2LS . CN‘NFCOCC CCIUCP‘NBFBAN NNCNCNN8N‘C01NCNN 5 3 00000000090040000 NNNNKNNNNNNNNNNNN 00.000000090090000 E CC [CECEECEECEECC r. L I F. 0 A F. R 4 0 6 8 1. 0 7 C N 2 H U 0 .L 8 U C 0 C N I 8 3. 1 1 R . A o C E . N . r. o . 8 ...r.r.H.Lr.r.r.E A U U U..- U U U U U DI. NNNNNENNNNNNNNNNN 0 El IRIRINRIRIRIRIRIR 0 A8 8UTU88J808U808J80 A3..53 ~8888N88|N8N8NTNTINTD cq ER OEOEOFEOEOCOCOEOEN 0 A8 A CRCRCIRCRCRCRCRCRC RLCNLC A0 AA 3 A 5 C 9 0 1 2 CRETIK 2 1. 1 1. TLSTN 9 CARD APPENDIX C SUGGESTIONS FOR FUTURE WORK 165 There are four areas of research that are suggested from these studies. 1. Ion Pairing of Complexes by Infrared Spectroscopy. The observation of the ion paired complex Na+-18C6, NCS- in tetrahydrofuran solutions by infrared spectroscopy (Chapter 3) appears to be the first such observation by this technique. It would be interesting to use infrared spectroscopy to investigate ionic association of complexed metal ions and compare these results with those of other techniques. 2. NaAlEt4 Complexation. This salt is soluble in a wide variety of low dielectric, low donor number solvents. In addition, the proton nmr of the anion provides informa- tion as to the type of ion pair the salt forms in a given solution (57). It would be interesting, therefore, to study ion pairing, complexation, and complexation kinetics of this salt in those solvents in which solubility problems prevent the use of the "normal" salts. 3. Cesium Ion Complexation Kinetics. It would be useful to determine if the dissociative exchange mechanism is pre- dominant for Cs+ complexation with DB3OClO in AC and in MeOH solutions as reported by Shamsipur (30). This 166 167 knowledge would complete the study initiated by Shamsipur concerning the kinetics of complexation of Cs+ ion with the larger crowns in these solvents. Mei EE.El° (28) assumed the dissociative mechanism to be predominant for Cs+ ion complexation with 18C6 in pyri- dine solutions. A free energy of activation of A61!2 = 12.02 kcal-mol-l was reported. However, if one assumes the bimolecular mechanism to be predominant, a value of 1 AG# = 9.3 kcal-mol- is calculated from their data. This 1 value is of the approximate magnitude one would expect for the solvent independent free energy barrier for alkali metal ion complexation with 18C6 based on charge density arguments. The barrier varies from mlO.5 kcal-mol-l for + 1 Na ion to m10.0 kcal-mol-l for K+ ion to m9.3 kcal'mol- for Cs+ ion. One would expect the free energy barrier to be highest for Na+ ion in this series due to its higher charge density which would result in the greatest amount of charge-charge repulsion in the transition state. Therefore, it would be extremely interesting to deter- mine the mechanism of Cs+ ion complexation with 18C6 in several solvents in order to compare the results with the other alkali metal ions. 4. Lithium Ion Complexation Kinetics. To our knowledge no data exist in the literature concerning Li+ ion com— plexation kinetics with crown ethers. It would be interest- ing to determine the kinetics of complexation of this ion 168 with the crown ethers and compare the results with the other alkali metal ions. Thus far, only Na+ ion has been demonstrated to ex- change via the dissociative pathway. Since Li+ has a higher charge density than does Na+, one would expect the free energy barrier for the bimolecular mechanism to be higher for Li+ ion. Therefore, it seems reasonable for the dissociative mechanism to also occur for the ion. It would be interesting to see if the above predictions are correct. REFERENCES 10. 11. 12. 13. 14. 15. REFERENCES Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 2495. 9 ’b Pedersen, C. J. J. Am. Chem. Soc. 1967, 8 , 7017. Dietrich, B.; Lehn, J.-M.; Sauvage, J. P. Tetrahedron Lett. 1969, 2885. Lehn, J.-M. Struct. Bonding (Berlin) 1973, 16, 1. Lehn, J.-M. Acc. Chem. Res. 1978, 11, 49. Liesegang, G. W. and Eyring, E. M. in "Synthetic Multidentate Macrocyclic Compounds" ed. by Izatt, R. M. and Christensen, J. J. Academic Press, New York, pp. 245-287 (1978). Schmidt, E.,Ph.D. Thesis, Michigan State University, East Lansing, MI, 1981. Szczygiel, P. M. S. Thesis, Michigan State University, East Lansing, MI, 1984. Wong, K. H.; Konizer, G.; Smid, J. J. Am. Chem. Soc. 1970, 2%, 666. Shchori, E.; Jagur-Grodzinski, J.; Luz, Z.; Shporer, M. J. Am. Chem. Soc. 1971, 3%, 7133. Chock, P. B. Proc. Nat. Acad. Sci USA 1972, 69, 1939. Liesegang, G. W.; Farrow, M. M.; Rodriguez, L. J.; Burnam, R. K.; Eyring, E. M. Int. J. Chem. Kin. 1978, K, 471. Rodriguez, L. J.; Liesegang, G. W.; White, R. D.; Farrow, M. M.; Purdie, N.; Eyring, E. M. J. Phys. Chem. 1977, 81, 2118. Farber, H.; Petrucci, S. J. Phys. Chem. 1981, fig, 1396. Chen, C. C.; Petrucci, S. J. Phys. Chem. 1982, 86, 2601. 169 16. 17. 18. 19. 20. 21. 22. 23. 26. 27. 28. 29. 30. 31. 32. 33. 170 Maynard, K. J.; Irish, D. E.; Eyring, E. M.; Petrucci, S. J. Phys. Chem. 1984, 83. 729. Chen, C.; Wallace, W.; Eyring, E.; Petrucci, S. J. Phys. Chem. 1984, 88, 2541. Boss, P., Ph.D. Thesis, Michigan State University, East Lansing, MI, 1985. Grell, E-z Funck, T.; Eggers, F. in "Membranes", ed. by G. Eisenman, vol. III, Dekker, New York, pp. 1-171 (1975). Gutmann, V., "Coordination Chemistry in Nonaqueous Solutions", New York, Springer-Verley (1968). Popov, A. I. Pure and Appl; Chem. 1975, 31, 275. Dickert, F. L.; Gumbrecht, W. Z. Naturforsch. 1984, 39b, 69. N'b’b Mei, E.; Popov, A. 1.; Dye, J. L. J. Am. Chem. Soc. 1977, 39, 6532. Kauffmann, E.; Dye, J. L.; Lehn, J.-M.; Popov, A. I. J. Am. Chem. Soc. 1980, 10%, 2274. Shchori, E.; Jagur-Grodzinski, J.; Shporer, M. J. Am. Chem. Soc. 1973, 2%, 3842. Shporer. M-; Luz, z. J. Am. Chem. Soc. 1975, 2;, 665. Schmidt, 3.; Popov, A. I. J. Am. Chem. Soc. 1983, 122, 1873. MGi: E-; POPOVI A- I.; Dye, J. L. J. Phys. Chem. 1977, $1, 1677. Mei, E.; Dye, J. L.; Popov, A. I. J. Am. Chem. Soc. 1977, 99, 5308. Shamsipur, M. Ph.D. Thesis, Michigan State University, East Lansing, MI, 1979. Lin, J. D.; Popov, A. I. J. Am. Chem. Soc. 1981, 10%, 3773. Day, M. C., University of Louisiana, Baton Rouge, Louisiana. Hohn, E. G.; Olander, J. A.; Day, M. C. J. Phys. Chem. 1969, 78, 3880. mm 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 171 Day, M. C. personal communication, 1984. Gokel, G. W.; Cram, D. J.; Liotta, C. L.; Harris, H. P.; Cook, F. L. Org. Sythn. 1977, 51, 30. Gokel, G. W.; Cram, D. J.; Liotta, C. L.; Harris, H. P.; Cook, F. L. J. Org. Chem. 1974, 32, 2445. Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 7017. Pedersen, C. J.; Frensdorf, H. K. Angew. Chem. Int. Ed. 1972, 11, 16. Wright, D. L. Ph.D. Thesis, Michigan State University, East Lansing, MI, 1974. Martin, M. L.; Delpuech, J(-J.; Martin, G. J. in "Practical NMR Spectroscopy", Heyden & Son Ltd., Philadelphia, p. 178 (1980). Templeman, G. L.; VanGeet, A. L. J. Am. Chem. Soc. 1970, 24, 5578. Greenberg, M. 5.; Popov, A. I. Spectrochim. Acta 1975, 31A, 697. (a) Van Geet, A. L. Anal. Chem. 1968, , 227. 0 m 2 m 4 W (b) Van Geet, A. L. Anal. Chem. 1970, 4 , 679. (c) Kaplan, M. L.; Bobey, F. A.; Chang, H. N. Anal. Chem. 1975, 41, 1703. Hertz, H. G.; Tutsch, R.; Versmold, H. Ber Bungenges Phys. Chem. 1971, 15, 1177. Eisenstadt, M.; Friedman, H. L. J. Chem. Phys. 1966, 44, 1407. Lindon, J. C.; Ferridge, A. G. Prog. in NMR Spectrosc. 1980, 13' 27. Walmsley, J. A.; Atkinson, T. V. unpublished program. Atkinson, T. V. unpublished program. Nicely, V. A.; Dye, J. L. J. Chem. Educ. 1971, 48, 443. Greenberg, M. 8.; Badner, R. L.; Popov, A. I. J. Phys. Chem. 1973, 77, 2449. ——--— 'Vb Carvajal, C.; Tolle, K. J.; Smid, J.; Szwarc,M. J. Am. Chem. Soc. 1965, 81, 5548. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 172 Comyn, J.; Dainton, F. S.; Ivin, K. J. Electrochim Acta 1963, 13, 1851. Bhattacharyya, D. M.; Lee, C. L.; Smid, J.; Szwarc,M. J. Phys. Chem. 1965, 69, 608. Tersag, G.; Boileau, S. J. Chim Phys. 1971, 68, 903. Boileau, S.; Hemery, P.; Justice, J.-C. J. Soln. Chem. 1975, 4, 873. Chabanel, M.; Wang, Z. J. Phys. Chem. 1984, 88, 1441. Ahmad, N.; Day, M. C. J. Am. Chem. Soc. 1977, 99, 941. Westmoreland, T. D. Jr.; Ahmad, N.; Day, M. C. J; Organometal. Chem. 1970, 25, 329. Vaes, J.; Chabanel, M.; Martin, M. L. J. Phys. Chem. 1978, 8%, 2420. Menard, C.; Wojtkowiak, B.; Chabanel, M. Bull. Soc. Chim. Belges 1972, 81, 241. Erlich, R. H.; Popov, A. I. J. Am. Chem. Soc. 1971, 9%, 5620. POpOV, A. I. in "Solute-Solvent Interactions" v.2, New York, Marcel Dekker, ed. by Coetzee, J. F. and Ritchie, C. D. (1976). Popovych, O. and Tomkins, R. P. T. in "Nonaqueous Solution Chemistry", New York, J. Wiley & Sons (1981). Frankel, L. S.; Langford, C. H.; Stengle, T. R. J; Phys. Chem. 1970, 74, 1376. Erlich, R. H.; Greenberg, M. S.; Popov, A. I. Spectro- chim. Acta 1973, 29A, 543. Kinsinger, J. B.; Tannahill, M. M.; Greenberg, M. S.; Popov, A. I. J. Phys. Chem. 1973, 11, 2444. Wu, Y.-C.; Friedman, H. L. J. Phys. Chem. 1966, 70, 501. Lin, J. D. Ph.D. Thesis, Michigan State University, East Lansing, MI, 1980. Frensdorff, H. K. J. Am. Chem. Soc. 1971, 9%, 600. Abragam, A. "The Principles of Nuclear Magnetism", Oxford University Press, London (1961). 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 173 Deverell, C. Prog. Nucl. Magn. Reson. Spectrosc. 1969, 4, 278. 'b Arnold, M. St. J.; Packer, K. J. Mol. Phys. 1966, 10, 141. Kintzinger, J.-P.; Lehn, J.-M. J. Am. Chem. Soc. 1974, 26' 3313. m Ceraso, J. M. Ph.D. Thesis, Michigan State University, East Lansing, MI, 1976. Gupta, R. K.; Pitner, T. P.; Wasylishen, R. J. Magn. Res. 1974, 13, 383. Woessner, D. E. J. Chem. Phys. 1961, 35, 41. Ceraso, J. M.; Smith, P. B.; Landers, J. 5.; Dye, J. L. J. Phys. Chem. 1977, 81, 760. Liesegang, G. W.; Farrow, M. M.; Vazquez, F. A.; Purdie, N.; Eyring, E. M. J. Am. Chem. Soc. 1977, 99, 3240. Cox, B. G.; Garcia-Rosas, J.; Schneider, H. J. Am. Chem. Soc. 1981, 10%, 1054. Shih, J. S.; Popov, A. I. Inorg. Nucl. Chem. Letts. 1977, 13, 105. Shamsipur, M.; Rounaghi, G.; Popov, A. I. J. Soln. Chem. 1980, 9, 701. (a) O'Reilly, D. E.; Petersen, E. M. J. Chem. Phys. 1969, 51, 4906. (b) Wehrli, F. W. J. Magn. Res. 1977, 25, 575. Mei, E. Ph.D. Thesis, Michigan State University, East Lansing, MI, 1977. DeWitte, W. J.; Schoening, R. C.; Popov, A. I. Inorg. Nucl. Chem. Letts. 1976, 12, 251. Khaezali, S. Ph.D. Thesis, Michigan State University, East Lansing, MI, 1982. "Nonaqueous Electrolytes Handbook“, Janz, G. J.; Tomkin, R. P. T. eds., Academic Press, vol. 1 (1972). (a) Allerhand, A.; Gutowsky, H. 8.; Jonas, J.; Meinzer, R. A. J. Am. Chem. Soc. 1966, 88, 3185. 174 87. (b) Allerhand, A.; Gutowsky, H. S. J. Chem. Phys. 1964, $1, 2115. "1111111111111411