NMR EXAMINATION 0F CYCLIC DIALKOXY 'CARBONIUM IONS (l, 3-DIOXOLENIUM CATIONS) Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY DONALD ANDREW TOMALIA 1968 LIBRARY Michigan State University THESIS This is to certify that the thesis entitled NMR ET’CANINATION OF CYCIJIC DL’XLKOXY CARTf‘NIUl-I ICNS (1.,S-DIOXC'Ll-JNIUI‘vi IONS presented by Donald Andrew Tomalia has been accepted towards fulfillment of the requirements for __Bh. Q ._ degree in flhemistry J Imd IIrf ‘ Major professor Date MaV 8, 1968 0-169 ABSTRACT mm EXAMINATION or CYCLIC DIAIKOXY CARBONIUM IONS (1, 3-DIOX0LENIUM IONS) by Donald Andrew Tomalia Members of the following four families of cyclic dialkoxy carbonium ions (1,3-dioxolenium cations) were prepared by allowing appropriate 2-bromoethyl esters to react with silver tetrafluordborate according to the methcd of Meerwein: I. 2-Alkyl-l,3-dioxolenium Cations II. 2,2'-Alkylene-l,3-dioxolenium Dications III. 2-Aryl-l,3-dioxolenium Cations IV. 2,2' and 2,2',2"—Aryl-l,3—dioxolenium Dications and Trications Families II and IV represent new examples in this series. An alternate method for thelneparation of 2-alkyl-l,3-dioxolenium nations was discovered which involved the combination of 2-hydroxy, methoxy or acetoxyethyl esters with an excess of fluorosulfonic acid (FSO3H). When 3H was added to the ester, l,3~dioxolenium.cations were generated im- xneiiately as the major product. PS“ Addition of the esters to FSOBH gave rise ‘to only small amounts of 1,3-dioxolenium cations accompanied by a predom- inance of diprotonatei ester species. The diprotonated species converted lemny'but completely to 1,3-dioxolenium cations with time. Evidence is gprwesented for the first example of protonation of the etheral oxygen in an ester. .A systematic rmu'examination of these four families of cations in ~liailli sulfur dioxide (-20°) or FSO3H revealed that the equivalent protons <3f tine dioxolenium.moiety could be used as a probe for assessing the Donald Andrew Tomalia electron density in the dioxolenium ring as a function of the 2-substitu- ent. A good qualitative correlation was obtained by comparing the chemical shifts of eleven 2-alkyl-l, 3-dioxolenium cations.2 In a similar manner, a good quantitative correlation of proton-nmr chemical shifts with Hammett O values was obtained for fifteen me_t_a_ and fi-substituted 2—aryl-l, 3-dioxolenium cations. This represents the first quantitative correlation, by proton magnetic resonance, of charge densities in a carbonium ion system.3 From this relationship and the chem- ical shifts of the 2,2'-n_t-phenylene and 2,2'-p-phenylene-l, 3-dioxolenium dications, Hammett C values for the m_e_t_§ and nag substituted dioxolenium moieties were found to be +0.8h and +0.97, respectively. The latter value is the largest positive 0 value reported to date. Application of this Hammett 0 relationship and the dioxolenium moiety probe for determining Hammett o values for higher energy carbonium ions is described. Nmr examination of 2,2'-alkylene-l, 3-dioxolenium dications showed that charge repulsion could be assessed as the number (n) of methylene insulating groups was varied.“ Dications containing five or six methylene insulating groups reflected practically total loss of charge repulsion and exhibited nmr chemical shifts which were reminiscent of mono-2-alkyl-l, 3- dioxolenium cations. By decreasing the number of methylene groups one 1”'Ound a smooth, monotonic increase in charge repulsion (as demonstrated by larger dioxolenium proton deshielding values). Deshielding was at a maxi- mm for n = l. Definitive evidence for the dication containing no insul- at1113 groups (i.e., n = O) was not obtained. An empirical relationship was conceived which correlated the nmr Chemical shifts of the dioxolenium ring protons in these dications with Donald Andrew Tomalia the number of methylene insulating groups (n) between the cationic centers (Equation 1). Alternatively, the chemical shifts of the alpha protons 6 = 5.30 + —l-'-§—0- (1) (ml)2 were related to the number of methylene groups (n) between the positive centers by the following equation: 5 = 2.9h (1 + 3%) (2) n The latter equation could be applied to other dicarbonium ion systems (e.g. acyl dications),5 possessing alpha protons, by choosing suitable parameters . REFERENCES l. H. Meerwein, V. Hederich and K. Wunderlich, Arch. Pharm., 221, 5&1 (1958)- 2. H. Hart and D. A. Tomalia, Tetrahedron Letters, g9, 3383 (1966). 3. D. A. Tomalia and H. Hart, ibid, g9, 3389 (1966). 1+. 3. Hart and D. A. Tomalia, ma, l2 13m (1967). 5. a. A. Olah and M. B. Comisarow, J. Am. Chem. Soc., 8_8, 3313 (1966). NMR EDCAMINATION or crcuc DIAncoxx CARBONIUM IONS (1, 3-DIOXOLENIUM CATIONS) By Donald Andrew Tomalia ATHESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1968 ACKNOWLEDGEMENTS The author wishes to express his sincere appreciation to Professor Harold Hart for his constant interest and guidance throughout this investigation. The author acknowledges indebtedness to The Dow Chemical Company and the Edgar C. Britton Research Laboratory for providing facilities and sponsorship of this work; special appreciation is extended to Dre. T. R. Norton and D. P. Sheetz for their inspiration and encouragement. To the author'swife, Elizabeth, a special note of appreciation is made for her patience and understanding. .The author also wishes to thanker. J. E. Schmidt for his assistance in typing this manuscript. ii I. II. III. TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . HISTORICAI’ O O O O O O O O O O O O O O O O O O O O O O O RESIMS O 0 O O O O O O O O C O O O O O O O O O O O I O A. B C. D 2'Aml- l, 3' diOXOlenim cations o o o o o o o o o o 2,2I-Alkylenebis-l,3-dioxolenium.Dications . . . . . 2‘A1'yl’l, 3" diOXOlenilm cat ions 0 o o o o o o o o o o 2,2I and 2,2',2"—Aryl-l,3-dioxolenium Dications and Trications O O O O O C O C O C O C O C C C O C C O O DISCIES'ION O O O C O O O O O O O O O 0 O O O O O O O O O 2'Aml' l, 3" diOXOlenim cat 10118 0 o o o o o o o o o 2,2'-Alkylenebis-l,3-dioxoleniwm Dications . . . . . Q'Aryl“ l, 3' diOXOlenium cat 10118 0 o o o o o o o o o o 2,2I and 2,2',2"-Aryl-l,3-dioxolenium Dications and Trications . . . . . . . . . . . . . . . . . . . . . MEMMAL O O O O O O O O O O O O O O O O O O 0 O O O A. General 0 O O O O O O O O O 0 O O 0 O O O O O O O O 'Melting Points . . . . . . . . . . . . . . . . . Microanalyses . . . . . . . . . . . . . . . . Nuclear Magnetic Resonance Spectra . . . . . . Infrared Spectra . . . . . . . . . . . . . . . . . Solvents . . . . . . . . . . . . . . . . . . . Miscellaneous . . . . . . . . . . . . . . . mm-P'WIUH Precursors to 2-Alkyl-l,3-dioxolenium.Cations . . . l. *2—Bromoethyl Acetate . . . . . . . . . . . . . . 2. 2-Bromoethy1 Acrylate . . . . . . . . . . . . . 3. 2-Bromoethyl.Methacrylate . 4. ‘2-Methoxyethyl Acetate . . . . . . . . 5. ~2-Hydroxyethyl Acetate . . . 6. 1,2-Diacetoxyethane . . . . . . . g. '2-Bromoethyl-N,N-diethyl Carbamic Acid Ester . . 9 . 2-Bromoethyl Cyclopropanecarboxylate . . . . . . 3,3-Dimethylacrylyl Chloride . . . . . . . . . . lO. 2-Bromoethyl-3,3-dimethylacrylate . . . . . . . ll. ngethoxycinnamoyl Chloride . . . . . . . . . . l2. 2-Bromoethyl-pfmethoxycinnamate . . . . ... . . l3. trans-Q—Bromoethyl Crotonate . . . . . . . . . . iii Page 20 21 hi 51 58 61 62 71 7h 11+. 15. 16. trans-2-Bromoethyl Cinnamate . . . . . . . . . . 2-Br0moethyl Pivalate o o o o o o o o o o o o o 2-Bromoethy1 Cycldbutanecarboxylate . . . . . . C. Precursors to 2-Aryl-l,3-dioxolenium Cations . . . . \£)(I>\l(7\\J'I-l="u)I'\)I--J 10. 11. 12. 13. 11+. 15. 16. 17. 18. D. Precursors to 2,2'-A1kyleneL.{1,3-dioxolenium Dications.‘ 2-Bromoethy1.prethoxybenzoate . . . . . . . . 2-Bromoethy1 3,h, 5- -Trimethoxybenzoate . . . . . 2-Bromoethyl prethbeenzoate . . . . . . . . . '2-Bromoethyl m—Methylbenzoate . . . . . . . . . ‘2-Br0m08thy1 Benzoate o o o o o o o o o o o o c 2'Br0moethyl EfFluorObenZOBte o o o o o o o o o '2-Bromoethyl pfchlorObenzoate . . . . . . 2-Chloroethyl.m¢ChlorObenzoate . . . . . . . . . 2-Bromoethy1 mrBromObenzoate . . . . . . . . . . 2-Bromoethyl meFluorObenzoate . . . . . . . . . '2-Bromoethy1 3,h-Dichlor0benzoate . . . . . . . 2-Bromoethyl merifluoromethylbenzoate . . . . . 2-Bromoethyl pfTrifluoromethylbenzoate . . . . . 2-Bromoethyl mrNitrObenzoate . . . . . . . . . . 2-Bromoethyl pritrObenzoate . . . . . . . . . '2—Bromoethyl Isophthalate . . . . . . . . . . . 2-Br0moethyl Terephthalate o o o o o o o o o o o 2-Bromoethyl Trimesate . . . . . . . . . . . . . 1. =Bia(2-bromoethy1) Oxalate . . . . . . . . . . . 2. QO'tUt-F’Uo 'Bis(2-bromoethy1)'Malonate .. . . . . . . . . . . Bis(2-bromoethyl; Succinate .. . . . . . . . . . 'Bis(2 -bromoethy1 Glutarate . . . . . . . . . . ' Bis(2-bromoethyl) Adipate . . . . . . . . . . . Bis(2-bromoethy1) Pimelate . . . . . . . . . . . BiB(2-br0moethyl) SUberate o o o o o o o o o o o E. 2-Alkyl-l,3-dioxolenium Cations . . . . . . . . . . 1. 2. 3 1. 5 6. 7 8 9 o 1 2- (N, N-Diethylamino) 1, 3-dioxolenium Tetrafluoroborate . . . . . . . . . . . 2eMethylpropenyl)-l, 3- dioxolenium 2-(prethoxystyry1)- 1,3-dioxoleniwm Tetrafluordborate . . . . . . . . . . . . trans-2-(Propenyl)- 1, 3-dioxolenium Tetra- fluordborate . . . . . . . . . . . . . . 2- 2- t Butyl)- -1,3- -dioxolenium.Tetrafluordborate . 1v 2-éCyclopropyl)- -l, 3- dioxolenium.Tetrafluordborate. 2- ' Tetrafluordborate . . . . . . . . . . . . . . 2-§Styry1)-l, 3- dioxolenium.Tetrafluordborate . . . Cycldbutyl)— —1,3-dioxolenium.Tetrafluoroborate . 2-Methyl-l, 3-dioxolenium.tetrafluordborate . . . . 2-(Isopropeny1)-1,3-dioxolenium Cation . . . . . . Page 101 102 102 103 101+ 105 105 106 107 107 F. G. 11. 2- (Vinyl)-L 3—dioxolenium Cation . . . . . l2. 2iflydroxfifl,3-dioxolenium.Fluorosulfonate . 2"AIyl- l, 3" DiCXOlenilm cat 10118 a o o o o o o o o 1. 2- (p-Methoxyphenyl)-L 3-dioxolenium Tetra- fluordborate . . . . . . . . . . - . . 2. 2- (3, 1T, 5-Trimethoxyphenyl)-L 3- dioxolenium I Tetrafluordborate . . . . . . . . . . . . 2-(prMethylphenyl)-l,3-dioxolenium.Tetra- . fluorOborate . . . . . . . . . . . . . .' 2-(mrMethylphenyl)-l,3-dioxolenium.Tetra- fluordborate . . . . . . . . . . . . .' 2- Pheny1)- -l,3-dioxolenium.Tetrafluordborate f1uordborate . . . . . . . . . . . . . . 2-(pyChlorOphenyl)-l,3-dioxolenium.Tetra- ' fluorOborate . . . . . . . . . . . . . 2-(m7Chlorophenyl)-l,3-dioxolenium.Tetra- fluordborate . . . . . . . . . . . . . . 2- (m—Bromophenyl) -1, 3- dioxolenium Tetra- fluordborate . . . . . . . . . . . . . 10.' 2-Qm7Fluorophenyl)-l,3-dioxolenium.Tetra- ' fluordborate . . . . . . . . . . . . . 3 h 5 6. 2- -Fluorophenyl)- -1,3-dioxolenium.Tetra- 7 8 9 11. _‘ 2- ( 3, I-I-Dichlorophenyl)-l, 3- dioxolenium Tetra- fluoroborate . . . . . . . . . . . . . 12. 2- (m-Trifluoromethylphenyl)-L 3- dioxolenium Tetrafluordborate . . . . . . . . . . . 13.. 2-(pfTrifluoromethylphenyl)-l,3-dioxolenium Tetrafluordborate . . . . . . . . . . . 11+. 2- (m-Nitrophenyl)- 1, 3-dioxolenium Tetra- ' fluordborate . . . . . . . . . . . 15. 2-(pyNitrophenyl)- l, 3-dioxolenium.Tetra- fluorOborate . . . . . . . . . . . . . 2'-Alky1ene-l,3-dioxolenium.Dications . . . . 3 l. Attempted Synthesis of 2,2'-Bi-l,3—dioxoleniwm Dication . . . . . . . . . . . . . . . Ea) With Silver Tetrafluordborate . . . . b) With Silver Hexafluoroantimonate . . 2. 2,2’-Methylenebis-l,3-dioxolenium.Tetra- fluoroborate . . . . . . . . . . . . 3., 2,2'-Ethylenebis-l,3-dioxolenium.Tetra- fluordborate . . . . . . . . . . . . h. 2,2'-Trimethylenebis-l,3-dioxolenium.Tetra- fluoroborate . . . . . . . . . . . . . 5. '- 2, 2' -Tetramethylenebis-L 3- dioxolenium Tetra- fluoroborate . . . . . . . . . . . . . Page 108 109 109 109 110 111 112 113 113 lit 115 115 116 117 118 118 119 120 120 120 120 121 121 122 123 VI. 3. 2,2'-Pentamethylenebis-l,3-dioxolentmm TetrafluorOborate . . . . . . . . . . . . 2,2'-Hexamethylenebis-l,3-dioxoleniwm Tetrafluordborate . . . . . . . . . . . . and 2,2',2"-Ary1-l,3-dioxolenium.Dications and Trications . . . . . . . . . . . . . . . 2,2'-m¢Phenylenebis-l,3-dioxolenium.Tetra- fluorOborate . . . . . . . . . . . . . 2,2'-p-Pheny1enebis-l,3-dioxolenium.Tetra- fluordborate . . . . . . . . . . . . . . 2,2',2"-Phenylenetris-l,3-dioxolenium Tetra- fluordborate . . . . . . . . . . . . . . Spectrum Spectrum. Spectrum ’ Spectrum ' Spectrum Spectrum ‘ Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum Spectrum. Spectrum Spectrum Spectrum Spectrum Spectrum ‘ Spectrum Spectrum Spectrum ' Spectrum Spectrum Spectrum ' Spectrum Spectrum 1. 21. 22. 23. 21. 25. 26. 27. 28. 29. 3o. 2-Bromoethyl-N,N-diethyl Carbamic ACid Eater o o o o o o o o o 2-Bromoethyl Cyclopropanecar- bOWlate O O O O O O O O O O 2-Bromoethyl-3,3-dimethy1acrylate trans-2-Bromoethyl Crotonate . . . trans-2-Bromoethyl;p-Methoxy- cinnamate . . . . . . . . . . 2-Bromoethyl Methacrylate . . . 2-Bromoethyl Acrylate . . . . . _ Bis(2-bromoethyl) Oxalate . . . Bis(2-bromoethyl) Malonate . . Bis(2-bromoethyl) Succinate . . Bis(2-bromoethyl) Glutarate . . Bis(2-bromoethyl) Adepate . . . BisE2-bromoethyl) Pimelate . Bis 2-bromoethy1) suberate . . 2-Bromoethyl prMethoxybenzoate NUCLEAR MAGNETIC RESONANCE SPECTRA . . . . . . . . . . . trans-2-Bromoethy1 Cinnamate . . . 2-Bromoethy1 Cyclobutanecarboxylate 2-Bromoethy1 Pivalate . . . . . 2-Bromoethyl Acetate . . . . . 2-Bromoethyl 3, 1, S-Trimethoxyben- zoate . . . . . . . . . . . . 2-Bromoethyl ngethylbenzoate . 2—Bromoethyl.mrMethylbenzoate . 2-Bromoethyl Benzoate . . . . . 2-Bromoethyl pyFluorObenzoate . 2-Bromoethy1 p:Chlordbenzoate . 2-Chloroethyl mechlorobenzoate 2-Bromoethyl m—Bromobenzoate . 2-Bromoethy1.meFluorObenzoate . 2-Bromoethyl 3,1-Dichlordbenzoate . 2-Bromoethyl merifluoromethyl- benzoate . . . . . . . . . . vi Page 123 121 125 125 125 126 127 128 129 130 131 132 133 131 135 136 137 138 139 110 111 112 113 111 115 116 117 118 119 150 151 152 153 151 155 156 157 Spectrum.3l. Spectrum 32. , Spectrum 33. Spectrum 31. Spectrum 35. Spectrum 36. - Spectrum 37. , Spectrum.38. Spectrum.39. ‘ Spectrum 10. Spectrum 11. Spectrum 12. I Spectrum 13. Spectrum.11. Spectrum 15. Spectrum 16. Spectrum I+7. Spectrum 18. spectrum 19. Spectrum.50. I Spectrum 51. l Spectrum.52. ‘ Spectrum 53. Spectrum.51. Spectrum.55. Spectrum 56. I Spectrum.57. Spectrum.58. 2-Bromoethyl pyTrifluoromethyl- benzoate . . . . . . . . . . 2-Bromoethyl mrNitrObenzoate . . 2-Bromoethyl p-NitrObenzoate . . 2-Bromoethyl Isophthalate . . . . 2-Bromoethyl Terephthalate . . . 2-Bromoethyl Trimesate . . . . . 2- (N, N— Diethylamino)-L 3- dioxolen- ium.TetrafluorOborate . . . . . 2-(Cyclopropyl)-1,3-dioxoleniwm Tetrafluoroborate . . . . . . 2-(2-Methylpropeny1)-l,3-dioxolene ium.Tetrafluordborate . . . . . trans-2-(Propeny1)-l,3-Dioxolenium Tetrafluordborate . . . . . . . 2- (p-Methowstyryl) -1, 3- dioxolenium Tetrafluoroborate . . . . . . . 2-(Styryl)-l,3-dioxolenium.Tetra- fluordborate . . . . . . . . . 2-(Cyclobuty1)-l,3-dioxolenium Tetrafluordborate . . . . . . . 2- (t-Butyl) -1, 3- dioxolenium Tetrafluordborate . . . . . . . 2- (Methy1)-L 3-dioxolen1um TetrafluorOborate . . . . . . . 2- (Isopropenyl) -1, 3- dioxolenium Tetrafluoroborate . . . . . . . 2- (Vinyl) -1, 3- dioxolenium Tetrafluordborate . . . . . . . Bis(2-bromoethyl) Oxalate and Two Equivalents of AngF . . . . . 2, 2' -Methylenebis- 1, 3- ioxolenium Tetrafluoroborate . . . . . . . 2,2'-Ethylenebis-l,3-dioxolenium Tetrafluordborate . . . . . . . 2,2'-Trimethylenebis-l,3-dioxol- enium.Tetrafluordborate . . . . 2,2'-Tetramethylenebis-l,3-diox- olenium.Tetrafluordborate . . . 2,2'-Pentamethylenebis-1,3-dioxe olenium Tetrafluoroborate . . . 2,2'-Hexamethylenebis-1,3-dioxr olenium.Tetrafluordborate . . . 2- (p-Methoxyphenyl) -1, 3- dioxolen- ium Tetrafluordborate . . . . . 2- ( 3, 1, 5-Trimethoxyphenyl)-L 3- Dioxolenium.Tetrafluordborate . 2- (p-Methyipheny1)-1, 3- dioxolen- ium Tetrafluoroborate . . . . . 2- (m-Methylphenyl)-L 3- dioxolen- ium.Tetraf1uoroborate . . . . . vii Page 158 159 160 161 162 163 161 165 166 167 168 169 170 171 172 173 171 175 176 177 178 179 180 181 182 183 181 185 Spectrum.59. Spectrum.60. Spectrum 61. Spectrum 62. Spectrum.63. Spectrum 61. Spectrum.65. Spectrum.66. Spectrum.67. Spectrum.68. Spectrum.69. Spectrum 70. I Spectrum 71. Spectrum.72. LITERATURE CITED . 2-(Phenyl)-l,3-Dioxolenium Tetrafluordborate . . . . . . . . 2- (p-FluorOpheny1)-1, 3- dioxolenium Tetrafluoroborate . . . . . . . . 2- (p-Chioropheny1)- 1, 3-dioxolenium Tetrafluordborate . . . . . . 2-(erhloropheny1)- -1,3-dioxolenium Tetrafluordborate . . . . . . . . 2- (m-BromOpheny1)- l, 3-dioxolenium Tetrafluordborate . . . . . . . 2- (m-Fluorophenyl)- 1, 3-dioxolenium Tetrafluordborate . . . . . . . 2- (3, 1-D1ch16ropheny1)- -1, 3-dioxo- lenium.Tetraf1uordborate . . . . 2- (m-Trifluoromethylphenyl)-L 3- dioxolenium.Tetrafluoroborate . . 2-(prrifluoromethylphenyl)-l,3- dioxolenium.Tetrafluoroborate . . 2- (m-Nitrophenyl)-l, 3- dioxolenium Tetrafluordborate . . . . . . . 2- (p-Nitropheny1)-1, 3-dioxolenium Tetrafluoroborate . . . . . . . . 2,2'-m-Phenylenebis-L 3- dioxolenium Tetrafluordborate . . . . . . . . 2,2'-prhenylenebis-l,3-dioxolenium Tetrafluoroborate . . . . . . . . 2,2',2"-Phenylenetris-l,3-dioxol- enium Tetrafluoroborate . . . . . viii Page 186 187 188 189 190 191 192 193 191 195 196 197 198 199 200 Table II III VI VII VIII LIST OF TABLES 2-Alkyl-l,3-dioxolenium.Tetra 2-Bromoethyl.Alkyl Esters . . Comparison of Proton Chemical Shifts (6) of 2-Bromo- ethyl Alkyl Ester Precursors (CClh) to 2-Alkyl- l,3-dioxolenium.Cations . fluordborates . 2,2'-Alkylenebis-1,3-dioxolenium.Dications . Bis(2-bromoethyl) Esters . . 2-Ary1-1,3-dioxolenium Cations . . . . . . . 2-Bromoethyl Benzoates . . Comparison of Proton Chemical Shifts (6) of 2-Aryl-l,3- dioxolanes (0011,) with 2-Aryl-L3-dioxolenium Cations . . . . . . . . Comparison of Proton Chemical Shifts (5) of 2-Bromo- ethyl Benzoate Precursors (0011,) to 2-Aryl-l,3- dioxolenium.Cations . . . . . . . . . . . . . . . . Nmr Chemical Shifts for 2-Aryl-l,3-dioxolenium. Multications in FSO3H . . Comparison of Proton Chemical Shifts of 2-Substituents ) with those in 2-Alkyl- in Ester Precursors (CCl 1,3-dioxolenium Cations 1 ix FSO3H).......... Page 22 21 26 13 16 52 51 55 57 59 66a Figure II. III. VI. VII. VIII. LIST OF FIGURES Reaction of 2-Hydroxyethyl Acetate with FSO H (Addition of Acid to Ester; Reaction Time = 15 Minutes) . . . Reaction of 2-Hydroxyethyl Acetate with FSO3H (Addition of Ester to Acid; Reaction Time = 7 Minutes) . . . Reaction of 2-Acetoxyethyl Acetate with FSO3H (Addition of'Acid to Ester; Reaction Time = 65 Minutes) . . . Reaction of 2-Acetoxyethyl Acetate with FSO H (Addition of Ester to Acid; Reaction Time = 15 Minutes) . . . Reaction of 2-Methoxyethy1acetate with FSO H (Addition of Acid to Ester; Reaction Time = 10 M nutes) . . Reaction of 2-Methoxyethyl Acetate with FSO3H (Addition of Ester to Acid; Reaction Time = 10 Minutes) . . . 2,2'-Alky1enebis-L 3-dioxolenium Dications in Order of Increased Shielding of Ring Protons . . . . . . . . Dioxolenium Ring Proton'Chemical Shift vs. Number of Methylene Groups 2-Methylene Proton Chemical Shift vs. Number of Methylene Groups Dioxolenium Ring Proton Chemical Shift vs. 11811111181313 0' (FSO3H) o o o o o o o o o o o o o o o o o 2-Substituted-L 3-dioxolenium Cation Families vs. nmr Chemical Shifts of the Ring Protons . . . . . . 2-Alky1-1,3-dioxolenium Cations . . . . . . . . . . . . Deshielding Parameters (A6) for Mono-, Di- and Tri-diox- olenium.Cations (FSO3H) Compared to their Ester Precursors (CClh) . . . . . . . . . . . . . . Page 30 32 31 37 10 1+7 1+9 1+9 578 63 79 INTRODUCTION Carbonium ions have been proposed as fleeting intermediates in many organic reactions since the early work of Mieerweinl and'Whitmore.2 Although the first example of a stable isolable carbonium.ion was re- ported by Seel in 1913,3 kinetic and stereochemical evidence for their intervention in reactions was common. The first use of nmr spectrosc0py as a means for characterizing carbonium ions was reported in 1958 by Doering and coworkers,u at which time nmr spectral evidence was presented for the heptamethylbenzenonium.ion. Since that time, techniques devel- ' oped primarily by Olah and Deno have made possible the direct observa- tion of many of these transient intermediates by nmr spectroscopy. Early progress in this area of carbonium.ion chemistry was reviewed in 1963 by 5a,b Deno and more recently by Olah.6 Since that time such a considerable volume of pertinent work has appeared in the literature that several texts have recently been published on the subject.7a’b At the present, alkyl,88-e cycloalkyl,9a-e benzyl,loa"e alkyny1,lla’b alkenylflea'f oxo,l3a-g fluoro,l)+a.c hydroxy,l§a"b alkoxy,l6a-d arylalkyl,l7a’b c 18a-c ions as well as bridged phenoniwm and benzenonium; at- 9a,b ions have been examined in some detail by nmr spectroscopy. These investigations have generally included attempts to qualitatively correlate proton.chemi- cal shifts with expected charge densities on the attached carbons. Al- though one successful quantitative correlation of fluorine nmr chemical shifts with stabilization energies has been reported for pgggffluorine substituted triphenylmethyl cations,20 in no instance has a quantitative correlation of charge densities with proton chemical shifts been demon- strated. In most cases the carbonium.ion structures were simply not amenable to correlation with linear free energy parameters, such as Hammett O values, but Olah has suggested that such a correlation.might be possible with the benzyl cation system.loe Conspicuously absent from the literature at the onset of our in- vestigation was an nmr spectral examination of alkoxy carbonium ions. Very recently Taft and Ramsey21 reported spectral data for a number of acyclic dialkow and trialkoxy carbonium ions of the type shown below; R—o . _ :IC-R X- + .: “" X R—o”' R09”, however, the cyclic dialkoxy carbonium.ions (2-substituted-l,3-dioxolen- 3:] x- ium cations) were considered of more interest for several reasons. A priori it was thought that the cyclic cation system might provide an in- teresting carbonium ion model, whereby the equivalent dioxolenium ring protons could be used as a prdbe for assessing charge delocalization or interaction with 2-substituents. It was also predicted that probe chem— ical shift deviations due to 2-substituent anisotropy effects would be at a minimum since the cyclic structure removes the probe protons from the immediate vicinity of the 2-substituent. Finally, using Meerwein's method22 one can introduce a wide variety of 2-substituents thus making possible a systematic and extensive investigation of charge density on the probe as a function of the 2-substituent. The investigation reported, herein, describes a systematic nmr spec- tral examination of these 2-substituted-1, 3-dioxolenium cations and pre- sents three new synthetic routes to these systems. Nmr data for 2- (m_ei_:_a_ and m-substituted aryl)-l,3-dioxolenium cations provide the first example of a quantitative correlation of carbonium ion charge density with proton nmr chemical shift as well as a unique and novel.method for the determination of Hammett o values for hydrolytically unstable moieties. Similarly a quantitative relationship between.proton chemical shift and charge separation was demonstrated for 2,2'-alkylene-bis-1,3- dioxolenium dications by using derived equations. The dioxolenium cations will be treated and presented as members of four distinct families of this series as described below: I. 2-Alkyl-l,3—dioxolenium Cations: an x- II. 2,2'-Alkylene-bis-1,3-dioxolenium.Dications: III. 2-Ary1-1,3-dioxolenium.Cations: 9 _ @{J x R 0 IV. 2,2' and 2,2',2"-Aryl-l,3—dioxolenium Dications and Trications: 3X“ II . HISTORICAL 2-Substituted—l, 3-dioxolenium cations were first postulated by Win- 23 stein in 1912 as transient intermediates to account for the exclusive formation of a trans-diacetate product from the reaction of trans-2-acetmy cycld'lexyl bromide with silver acetate. Similarly the acid catalyzed O —+ O ...-1 C 0 0‘7) CH3 I Br 0- 0- CH3 trans _. hydrolysis of cyclic ortho esters21+ was proposed to involve such a cation- ic intermediate yielding in this case a cis product. I' ' f '- o i»- 4 0 0.,+,0 ’1 9 o-d-CH (3’6 H H - 3 0:3>< OEt CH3 CH3 <1; £1.91 cis A survey of the literature makes it apparent that these cationic intermediates are involved in a wide variety of organic reactions. The numerous examples can be grouped in three general categories. A. Displacement of a kSubstituent by Participation of a Carboxy Group 0 II R-C-o-CH2-CH2-x -’—X—-> R-—CH3 0Ac 0 undergo any reaction under these conditions even after several days. It is interesting that 38, 5Ghdiacetoxy-6B-fluorocholestane converts .readily to the analogous six membered l,3-dioxenium.cation in perchloric acid even though the acetoxy groups have a trans relationship to each other.39 11 HC101+ ‘Iv C101+ AcO The first spectral characterization of a 1, 3-dioxolenium cation as an intermediate in a reaction was reported by Wilcox and Nealy in 1963.’+0 They found that when benzoate esters of either gig or grins-l, 2, 3,1—tetra- methy 1cyclobutene-3,1—diol were treated with 97% sulfuric acid or boron trifluoride these cationic species were generated and could be identified by nmr spectroscopy. Chemical proof of structure consisted of decomposing these cations in methanol and water to give an ortho ester and a cis- hydroxy benzoate, respectively. This criterion was employed earlier by Winstein and coworkers as a proof of structure.23’ 2"" CH CH3 0 CH CH3 ’ "_ H so I 2 1 A L.__VCH3 9. BF3 7 I ”a o—c- c CH3 9’ A CH30H/ 3 l CH CH3 / CH CH -. 3 "0 911300)< 0CH3 I H3) (pH CH3 ¢ CH3 i ¢ CH CH 3+) 320 ,_ 3 «03110 ’ I 7‘3 *1 ,¢ 12 1,3-Dioxolenium.ion intermediates have been.proposed in the Br¢nsted and Lewis acid-catalyzed ring opening of vicinal epoxides (cyclic ethers) 11,12 13 bearing a neighboring trans-acetoxy group. Coxon and coworkers found that 3B-acetoxy-1a, 5a-epoxycholestane gives a stable 3B, I+B-bridged ionic complex with BF which upon hydrolysis gave the corresponding diol. 3’ BF 3 -——9 A00 ””6 1. 11,11 Buchanan has shown that a Brdnsted acid can cause a similar transformation, presumably via the l,3-dioxolenium.cation as shown.below: CIIQ-O-C-¢3 F GHQ-0H 1 F CHQ‘OH AcO 0 HCl 3 ; CH 3 (’ 0CH3 O-CH3 + -* H - I— H _ GHQ-O 0 C0Me2 .1 0 3 (r CH3 0H 35 In the case of acyclic ethers, Meerwein and coworkers were able to Prepare 2emethyl-l,3-dioxolenium.tetrafluorOborate in good yield by the reaction of 2-ethoxyethyl acetate with either triethyloxonium.tetrafluoro- borate or BF3 according to the following equations: l3 + ' I R (Et)30BFu- R CH3-C-0-CH2-CH2-O-Et > $3 -qb /Et _ -(Et)20 -CH20H2-0 13F,l BF3 L We) 8 BF3 CH - -Et CH ‘C ' 2 '0 _ (321% II / CH -<:O_1Et0BF3- 3 o__I -B(0Et)3 The latter reaction is extremely slow and generally requires 1/2-1 year for completion. Antimony pentaChloride reacts in a similar but more rapid manner to give a 56% yield of the above cation as a hexachloroantimonate- salt. In special cases where labile groups are located 8 to a carboxy group, participation occurs readily under solvolytic conditions to give products derived from l,3-dioxolenium cation intermediates. This aspect has been investigated in considerable detail by Winstein and coworkers.23’h5 They found that _t_I_'_a_n§-2-acetomcycloheiql brosylate solvolyzes in acetic acid to Egg-L 2-diacetoxycyclohexane, presumably via the l,3-dioxolenium cation, several hundred times faster than does gig-2-acetoxycyclohexyl 'brosylate. The enhanced rate is undoubtedly due to anchimeric assistance by the acetom group in the _t_r.a_n§ isomer, whereas solvolysis of the gig isomer is governed by the rate at which the substrate loses brosylate ion, since anchimeric assistance is geometrically not possible. 11 0 OAc ~ 0B8 OAc HO2 (XE +—— ‘1‘: More recently Schneider and Lang“6 examined the anchimeric effects of the benzoyloxy group in gig and m—2-benzoylonqcyclohe3ql tosylates in anhydrous acetic acid. The relative rates for the gig and m isomer were foun to be 1.6 x 104+ and 0.26 respectively, compared to cyclohexyl tosylate as 1.00. Several olefin oxidations have been reported in the past decade which are best explained in terms of 1, 3-dioxolenium ion intermediates. Brutcher and Vara’+7 found that cyclopentadiene undergoes a facile oxidation with lead tetraacetate to give the M diacetate in the presence of acetate anion, the gig hydroxyacetate in wet acetic acid and gig diacetate in anhydrous acetic acid. These authors interpret the above transformations 5 OAc OAc according to the following reaction scheme: :>-—-> '- fl 0H OAc OAc OAc 15 Grinstead1+8 postulated the intermediacy of l,3-dioxolenium.salts in the oxidation of olefins with thallium salts to hydroxyesters and diols. Later work by Winstein and Andersony9 on the stereochemistry of this reac- tion in acetic acid supported this conjecture. These workers envisioned the reaction mechanism.in the following manner: Tl(0Ac) 9 3 ’ .:T1(0Ac)2 __) (.>— on 7 OAc . 3 et Dry H OAc OAc O‘c OAc In anhydrous acetic acid the diacetate was mainly t£2n§_(up to 88%), ‘whereas in wet acetic acid the diacetate was primarily gig (up to 81%). The significant reversal of diacetate stereochemistry by water is consid- ered to be a criterion for the intermediacy of a l,3-dioxolenium cation. Recent work reported by Olson50 on the selenium dioxide oxidation of ethylene in acetic acid has implicated l,3-dioxolenium.cations as inter- mediates in these reactions. Alkoxycarbonium.ions have been suspected as intermediates in the acid- catalyzed hydrolysis of acetals, ketals and orthoesters for some time.51 1&1 the case of an orthoester a dialkoxycarbonium ion would be the expected intennnediate. Under normal reaction conditions their isolation was 16 O'R' ' H+ O-R , R—C—O-R' ———’ R<+ + ROH _R, O-R' $ Hydrolysis Products precluded in that nucle0philic species were usually present thus leading to their destruction. It was not until the pioneering work of Meerwein and coworkers that this speculation was soundly corrOborated. In an ex- tensive investigation.which began in 195552.Meerwein found that both acyclic22 and cyclic35 dialkoxycarboniwm ions could be prepared and iso- lated by treating appropriate orthoesters with an excess of a Lewis acid (BF3, SbCls) or a Brdhsted acid possessing weakly nucleophilic anions (i.e., HQSOA). An excess of these reagents was essential in order to complex any nucleophilic species which were generated in the reaction. By using only a catalytic amount of the acid reagents one merely observed ring opening to the corresponding 2-alkoxyethyl ester. These reactions led to the l,3-dioxolenium cations in high yield (60-97%), unless electron hBF3 , T _ o g) R—(fi 13F1+ + B-(OEt)3 E X“ 28bCl o 2H250h , 1 - R (SJ SbCl6 + EtOSbCJh Catalytic Amount of BF ‘ 3 . <3. 4:3 l - R—C-O-CHe-CH2-0Et R \+ 3501+ + Etoso3H + H20 withdrawing groups are present in the 2 or M position. For example, BF3 Will not convert the following dioxolanes to the cations, but prefers to remain as a complex, and SbCl5 transforms the h-substituted dioxolane to the corresponding dialkoxycarbonium.ion in only 21% yield. 17 {Gym [0: 0Et O R O 2-Cl According to recent work by Winstein, this was the method of choice Cl- 032 for the preparation and isolation of 2—methyl-cis-h,5-tetramethyleneol,3- 3h dioxolenium tetrafluoroborate. °:>§H3_3____’ BF '1??? 0 CH 0 3 l, 3-Dioxolenium cations have been prepared and isolated in several instances by the oxidation of 2-substituted-l, 3-dioxolanes. Meerwein52 first reported this general method in 1955 when he isolated the parent 1, 3- dioxolenium tetrafluoroborate in 62% yield by the reaction of l, 3-dioxolane 53 with trityl tetrafluoroborate. Meerwein provided subsequent variations __0 H _ _o>% + (gag-Earn —-> |:E>—H + ¢3—cs of this reaction by oxidizing 2-phenyl-l, 3-dioxolane to the corresponding cation with triethyloxonium tetrafluoroborate and also with a mixture of ethyl bromide and silver tetrafluoroborate. In the latter case, a very + - -Et20 o \ ’ 1+ BF ’ © "0 h c 2H5.Br, AgBF u/ -C2H6, AgBr transient primary carbonium presumably affects the oxidation. Dialkoxycarbonium ions have been implicated as intermediates in the 18 halogenation of acetals. Marvell and Joncichsh found that the bromination of benzaldehyde diethylacetal with N-bromosuccinimide proceeded smoothly to give a bromine-free product which was identified as ethyl benzoate. It was conjectured that this product arose in the following manner: /0Et-_9 ’40Et - © C<-H ———)©-c-Br: :2 C'.' + Br OEt V‘OEt + EtBr :O-Et Cyclic dialkoxycarbonium ions were suggested as transient intermed— \é—— iates in the chlorination and bromination of cyclic acetals. Cort and Pearson55 found that the halogenation of l, 3-dioxolane with bromine or chlorine gave the corresponding 2-bromo- or 2-chloroethyl formate directly. Chlorination of 2231.971: 1+, 5,8-tetraoxadecalin yielded a mixture of b__i§_-2- chloroethyl oxalate and 2, 3-dioxo-l, h-dioxane. These products were believed to have resulted from a dication as shown below: 0 I C (XXII—2» (STEM-e '8 o u C-0— (CH -Cl 2’2 0 O 568 Schmitz and coworkers proposed a l, 3-dioxolenium cation as an in- termediate in the oxidation of 2-phenyl-_c_i_§-h,5-tetramethyl-l,3-dioxolane with N-bromosuccinimide. This conjecture was based on the fact that 13929.3." 2-bromoalkylbenzoate was obtained exclusively and would be the expected Product from such a cationic intermediate, whereas a radical rearrangement might be expected to give a cis-trans mixture. ‘5‘: —“§“’-> 01(w More recently Prugh and McCarthy5&postulated a radical rearrangement mechanism to account for the formation of 2-bromoethyl esters from the reaction of NBS and catalytic amounts of 2, 2'-azobisisobutyronitrile (AIBN) with 2-substituted-l,3-dioxolanes. These authors could not provide a O p 0 II II R-C-O-CH -CH -Br :K: |____) AIBN R- <: I a R_C\ NBS; + 2 2 O-CH -CI'I° Succinimide radical plausible explanation, however, for the unusual ring opening aptitude of 2,1t-disubstituted-l, 3-dioxolanes under these conditions. 2-Phenyl-ll- methyl-l, 3-dioxolane gave a 92%. yield of an isomer mixture of (A) and (B) with NBS and AIBN. The isomer ratio of A : B, however, was found to be 5 : l and seemingly inconsistent with the proposed radical rearrangement CH3 0 C + C. (EH3 AIBN \ O- CH-CH 2- Br \ 0- CH2- CH-Br I (B) mechanism, which would require preferred rearrangement of a dialkoxybenzyl radical to a primary methylene radical rather than a secondary radical. In ' View of the evidence56’ 57 supporting Sn2-like ring opening reactions of l, 3-dioxolenium cations the above product distribution may be best ration- alized via such a cationic intermediate. III RESUME 21 A. 2-Alkyl-l,3-Dioxolenium Cations: The 2-alkyl-l,3-dioxoleniwm cations used in this investigation were prepared according to a modified version of the.Meerwein.method.22 By allowing equimolar amounts of 24bromoethyl esters and anhydrous silver tetrafluoroborate to react in methylene chloride at 25-30° for 1-5 hours, l,3—dioxolenium tetrafluordborates were generally Obtained as nice white isolable salts in yields of h3-93% (see Table I). Interestingly, 2-(27 methoxystyryl)-l,3-dioxolenium.tetrafluoroborate, which was brilliant cans ary yellow, was the only colored salt ObserVed in this entire series and will be commented on later. It was necessary to use distilled or purified ester precursors for the cations or low yields and inferior products invar- iably resulted. Conversion of the esters to l,3-dioxolenium cations was 0 Alkyl-g-O-CH -CH H-Br 118—EL AMI-<3] BF); 2 2 CH2012 + AgBr unambiguously ascertained by observing the disappearance of the A2X2 nmr pattern characteristic for the esters and the formation of a sharp singlet for the equivalent ring methylene protons of the cations. The 2-alkyl-l,3-dioxolenium.tetrafluoroborates were generally soluble in methylene chloride with the exception of several which contained a double 'bond as part of the 2-substituent, [i.e., 2-vinyl, 2-isopropenyl, transf2- .Perenyl, 2-styryl and 2-(pfmethoxystyryl)]. .All of the 2-alkyl cation salts were soluble in acetonitrile, liquid sulihxr dioxide or fluorosulfonic acid. The last two solvents served as excellent media for nmr analysis of these cations. fEhese salts were extremely moisture sensitive; several examples have been reported 57 to undergo facile ring opening with water to produce 22 TABLEI 2-Allqyl- l, 3-dioxolenium Tetrafluoroborates R—é] BFh- Reaction ' Nmr I; g9, °C. % Yield Time, Hr. §pectrum 1. Et2N- 53-5h.5 69 1 37 2. D- 121+. 5-126 66 2 38 3. (CH3)2C=CH- 61-62 65 2 39 1.. CH3-0-©-CH=CH- 220-223 71 1 1+0 5. CB3-CH=CH- 150-152 79 1 1+1 6. ©-cn=cn- 178-179.5 67 1.75 1+2 7. 0- 31-32 56 1 M3 8. (CH3)3C- 151.5-152.5 72 2 1+1!- 9. CH3- 170-172 83 1 1+5 $33 10. CH2=C- 155-156. 5 81 2 1+6 11. CH =CH- 151-152. 5 #3 1+.5 1+7 23 2-hydroxyethyl esters. For this reason all operations were carried out in a glove box under scrupulously dry conditions. Under anhydrous storage con- .0 _ H20 " (+ BF ----> RPC'O‘CH -CH -0H + HBF \0 )4- 2 2 LI- ditions the purified salts appear to have long shelf lives (i.e., 2 years) whereas impure samples tend to deteriorate within several weeks to a dark, amorphous mass. The ester precursors were readily prepared in yields of 25-87% by re- fluxing equivalent amounts of 2-bromoethanol with the appropriate acid chloride in carbon tetrachloride for 3-2h hours (see Table II). These 2- bromoethyl esters were generally obtained as distillable, colorless liquids or as crystallizable solids. The esters exhibited expected carbonyl ab- sorptions in the 1700-18000m-l region as well as giving appropriate ele- mental analyses and nmr spectra (see Spectra l-ll). Nmr spectra of the cations were obtained both in liquid sulfur dioxide (-20?) and in fluorosulfonic acid (FSO3H) (see Spectra 37-h7). A downfield solvent shift of approximately 0.16 ppm.was observed in going from FSO H to 3 SO Tetramethylsilane (TMS) was used as the internal standard in sulfur 2. dioxide; in FSO H, the reference was tetramethylammonium.tetrafluoroborate 3 (TMA-BFA). The latter compound was assumed to have an absorption peak at ‘3-lO ppm.downfield from tetramethylsilane, the value reported for this material in 100% HQSOA.58 Compared to several 2—alkyl-l,3-dioxolanes, the ring protons in the corresponding cations were generally shifted downfield l-# to 1.7 ppm in FSO3H. For example 2-methyl—l,3—dioxolane had the follow- ing chemical shifts (0) in CClu compared to 2-methyl-l,3-dioxolenium tetrafluordborate in FSO3H: (CH3)3C- CH- CH CH =C- CH =CH- 2h TABLE II 2—Bromoethyl Alkyl Esters 0 II R- C- O- CH 2- CH 2- Br or B 77-79°/1.5mm 81-82°/10mm 78-79°/hmm 50-5-51-5 llu-115° /lIOmm uu.5-uo 79- 80° /5mm 9h-95°/3hmm 160- 161° /7l+0mm 1+6— 1+9° /5mm 52-53°/5mm % Yield Reaction timez hr. 56 20 87 5 73 10 81+ 3 50 h 62 2 2h 77 15 2 5 10 (Eastman Kodak) (Borden Company) (Borden Company) Spectrum 10 w 9‘ I p ...: 5K v 4—» :x ...—b 25 501 . l‘fll . l—l. 28(d) l—2. 75(8) "h.87(q) A0=1J+7 A0: ".1 2+8 In Table III the proton chemical shifts of the 1, 3-dioxolenium cations are listed in order of increased deshielding of the ring protons (c) and are compared to the methylene groups .(a) and (b) in the corresponding ester precursors. In FSOBH the Ad's for (b) and (c) varied between 0.65-0.95 ppm, whereas in liquid sulfur dioxide A0 varied between 0.95-1.13 ppm. The trend toward larger Aé's in going from cation 1 -’ cation 11 in FSO3H sug- gests that the more shielded cations possess 2-substituents which are more effective for delocalization of positive charge and provides at least a qualitative basis for using the cation ring proton chemical shifts as an electron density probe. This criterion has been employed by Olah6 and Deno5 a, b in other carbonium ion systems and has been found to provide a quantitative relationship for the 2-aryl-1, 3-dioxolenium ions, which will be discussed later. Whereas 2- (p—methoxystyryl)-1,3-dioxolenium fluoroborate gave a bril- liant yellow solution and a simple, explicable nmr spectrum in liquid sul- fur dioxide, in F80 H the cation was immediately decolorized and gave a 3 somewhat more complex spectrum. In the acid medium two groups of appropri- ate but displaced resonance signals were observed for the cation. This is interpreted as being due to partial protonation of the methoxy group to PIOduce a mixture of the monocation and a dication. The chemical shifts of these two species are shown below: 26 TABLE III Comparison of Proton Chemical Shifts (0) of 2-Bromoethyl Alkyl Ester Precursors (0011*) to 2-Alkyl-l, 3-Dioxolenium Cations o (c): O 1’ "' .. (b) (a) HQ x B R—C-O-CHQ-CHzBr (C). (CClh) (FSO3H,25°) \(302,-20°) (a) (b) (c) A0 (c)' A0 lo EWEN- 3.53 h.33 h.98 0.65 ——- -—— 2. [:>»— 3.50 4.35 5.17. 0.82 5.38 1.03 3. (CH3)2C=CH- 3.h8 1.35 5.19 0.8h -— -- h. CH -0- CH=CH- 3.52 h.hh 5.22 0.78 ‘-— -—— 3 Q (5.29)* 5. CH3-CH=CH- 3.52 1.10 5.23 0.83 5.38 0.98 6- ©-CH=CH— 3.53 1+.l+6 5.25 0.79 5.111 0.95 7- <:::>c_ 3.19 h.36 5.29 0.93 5.15 1.09 8- (CH3)30- 3.50 1.35 5.30 0.95 5.u8 1.13 9- CH3- 3.50 n.35 5.30 0.95 5.11 1.09 3H3 10. CH2=C- 3.51» 1.1.3 5.31. 0.91 5.52 1.09 ll" CH2=CHL- 3.52 4.11 5.35 0.91 5.52 1.08 *Chemical shift for the methoxy protonated cation I; H CH3. © /H ’,0 0113- © C/< ‘C0 + FSO3H ———+ E>fl0—H F803- ppm singlet) were deshielded more than those in the cyclic dialkoxycarbonium ions 1-6 (Table III), even though one might expect considerable charge delocalization through the resonance structure shown above. A similar obser- 112,21 vation was reported by Taft and Ramsey for the analogous acyclic series. They found that the methoxy protons were more deshielded in the trimethoxy cation than in the dimethoxy cation shown below. A good explanation for CH3 0A0 é I H3 CH3 0 = 5.0 ppm 0 = 5.1 ppm these observations is not immediately obvious, unless the ethylene carbonate is being protonated on the ether oxygens. In this investigation, three new methods were discovered for the gen- eration of simple 2-substituted—l,3-dioxolenium cations. It was found that treatment of either a 2-hydroxyethyl, 2-acetoxyethyl or 2-methoxyethyl ester with an excess of FSO H resulted in the formation of the dioxolenium 3 V v. ~\b .Ysu 28 cation and associated protonated species, these depending on which precur- sor was used. It was fUrther found that the mode of combination of these reactants was very important. Certain interesting protonated ester species could be generated in preference to the l,3-dioxolenium cations by adding the ester to an excess of FSOBH. These protonated ester species were grad- ually converted on standing (i.e. up to 5 months) to the corresponding 1,3- dioxolenium cations. In contrast, by adding the acid to the ester, the l,3-dioxolenium.cations were observed to form.immediate1y as the major species present. 2-Hydroxyethyl esters were the best precursors to the cyclic cations. By adding FSO H to 2-hydroxyethyl acetate, 2amethyl-1,3-dioxolenium.fluoro- 3 sulfonate was generated almost quantitatively within #5 minutes (see Fig- ure I). Reversing the order of addition (i.e. adding ester to acid) gave predominately the diprotonated ester (A) which was then gradually converted to the l,3-dioxolenium.cation upon standing at room temperature for several days (see Figure II). These transfonmations might be represented in the following manner: 0 /\FSO H 0-H h , CH3-C-0-CH2-CH2-0H 3 > CH3-C.<+ + 1.3 0-CH -CH -0- 2 2 H (A) -..—N 0 +H+ lT-H+ / “ 0 , CH3-C-0-CH2-CH2-0H .7 FSOBH : CH3-C + /H ‘o-CH -CH -0‘ 2 2 \H (B) M k + . {‘0 - - CH3-§;] FSO3 + H3OF'SO3 29 lignre I. Reaction.of 2aflydroxyethy1.hcetate‘with 15031: (Addition of Acid to Ester; Reaction‘rflle - hSIIinutes) 0 . c) “rig-Wren.“ 3035-: 7.) «33:21) "”3" (c) Uhknawn.(b) (m (a) Assiggggnts (0) x 3.10 (Ills-31h) a 2.75 (I) b h.11 (s) c 5-31 (0) 30 Figure 11. Reaction.of’zanydroxyethyl.hcetate with I803fl (Addition.of’lster to.Acid; ' Reactionfirmlsi- 7ilinutes) on + A ‘mmz-cnz—5\n (c) 23803 d) '(a) 033 10 (d;fllh;' ‘3 m (b) ca,-( cna-c-o-cse-me-os —L + (1:) —~ (d) (e) rs, aim-9.2m x 3.10 (ma-ark) a 2.75 (I) b '2.77 (c) 6 M91 (1) 4 5-31 (0) («ah (ah) - 11:3 31 2—Methy1-l,3-dioxolenium cation (Figure I) was identified unequivo- cally by comparison to an authentic sample which had been.prepared accord- ing to the method of Meerwein.22 The identity of the singlet at h.7l ppm is not readily apparent at this time. It should be mentioned that this same singlet (0 = h.72 ppm) was observed in the conversion of 2-hydroxyethyl benzoate to 2-phenyl-l,3-dioxolenium fluorosulfonate. 2-Methyl-l,3-dioxolenium.cation appeared as the minor product in Fig- ure II with singlets at 2.75 and 5.31 ppm. The major product in Figure II was assigned the diprotonated structure (A). This assignment was consistent with the A2B2 pattern at h.9l ppm and the singlet at 2.77 ppm. Olah re- ported60 a value of 2.75 ppm for the corresponding methyl group in.pro- tonated ethyl acetate. The diprotonated species (A) may be longer lived than (B) because the carbonyl group is no longer effective as a nucleophile for backside dis- placement in this dicationic form, Protonated ester (B), however, can liberate water by intramolecular displacement via the carbonyl group to yield the cyclic cation. The facile conversion of (B) to the cyclic cation presumably precludes Observation of (B) under these conditions. Dication (A) was Observed to slowly convert to the cyclic cation. This presumably ~CH3 CH3.<+ 1;! g: O-CHe-CH2—O O-CHE-CHQ-g-C-CH3 (C) (D) assigned to the unsymmetrical diprotonated ester (D). Some assurance for this speculation was gained by noting that the ratio of (g + f + e) : (a + b + c + d) was #:3, thus giving a proton ratio which was consistent with our assignments for (C) and (D). As the mixture stood at room.temper- ature, the singlet at 5.31 ppm due to the ring protons in the cyclic l,3- dioxolenium.cation increased with time, whereas the intensities of the singlet (5.12 ppm) and multiplet (1+.93 ppm) decreased a commemurate amount. These changes in intensities were followed in one instance and are as shown below: Intensity (e + :1 Reaction Time Intensity (3] 1 hr. 3 2 days 1.69 6 days 1.15 7 days 1.11 9 days 0-89 5 months 0 After 5 months the conversion is complete and quantitative to the cyclic cation and protonated acetic acid. After that time the spectrum.consists of only three singlets, located at 5.32, 2.8h and 2.76 ppm respectively. These spectral data may be rationalized according to the following reaction.scheme: 36 O O ,. FSO H H Ho ,0. Ho CH3- --0-C o—CH2 —2CH --0 CCH3 —3—> CH3+ 'CH3 14.». 033-0 \ + CH3 CH -CH2- -0 0 CH -CH - 2 H, 2 2 a - + W is O 8 H u ~ and H O 3 C‘o-CH -CH -§CH3 or CH3 C\O-CH -CH -()-g-CH 2 2 2 2 + 3 -HOAc OH CH3<;H F803 + CH3-<:+:| F803 Conversion of 2-methoxyethyl acetate to the cyclic cation by addition of acid to the ester appeared to parallel the previous two examples, in that this mode of combining the reactants also gave the cyclic cation as the major product accompanied by minor amounts of associated protonated species (see Figure V'). Resonance signals which were extraneous to those for the cyclic cation were as follows; multiplet at h.67 ppm, singlet at h.39 ppm, a multiplet at b.25 ppm.and a singlet at 2.70 ppm, By adding a small amount of methanol, the multiplet at h.25 ppm was enhanced; thus this peak is as- signed to protonated methanol.62 The signal at h.67 ppm.was assigned to the methylene protons in the diprotonated ester (E) whereas the singlets at it~39 and 2.70 ppm.were assigned to the ethereal and carbonyl methyl groups, respectively in the dication (E). Olah's work61 supports the as- Signment at h.39 ppm in that he reports a 6 value of h.39 ppm for the anal- Ogous methyl group in cation (F). Furthermore signals (e) and (d), Figure V: were present in a ratio of approximately h : 3 which is consistent with 37 Figure V . React ion of 2-Methowethyl Acetate with 1803!! (Addition of Acid to Eater; Reaction Time - llO Minutes) (a) CH 3-C 0 CH2 CH2 .9 CH3 -H+ o u ‘H o CH-C /\, I? 9251 CH3—<:+ A 3 \ o-CH -CH -o-CH -CH -CH -o-CH 2 £51- 3 V 2 2 3 -CH3OH -HOAc o-H + H .+ + - C 3 ] CH3-0 CH3 F303 F803 F803 +/H CH3-Q\ H F803 It should be noted that signals (e) and (d) might alternatively be assigned to the cyclic methyl oxonium cation shown above which could be formed by participation of the methoxy group with concurrent displacement 0f acetic acid. The reaction mixture resulting from addition of 2—methoxyethyl acetate 39 to FSO3H gave a considerably more complex spectrum (see Figure VI). 2- Methyl-l,3-dioxolenium cation and protonated methanol could be unequivo- cally identified as well as the diprotonated ester seen in Figure V. However, superimposed upon these signals were new resonance bands at 5.05 ppm (h) and at h.72 ppm (3). The identity of these signals was not readily apparent. These latter two resonance bands appeared to be decreasing in intensity with time (25°) as the l,3-dioxolenium cation signal (5.31 ppm) increased in intensity. Intrigued with our success in cyclizing the forementioned acetate esters, it was thought that it might be possible to ionize 2-bromoethyl acetate in FSO3H to produce l,3-dioxolenium.cations. Nmr analysis indi- cated that the ester merely underwent carbonyl protonation without loss of bromide ion. This was readily apparent by noting the large downfield shift 0 I FSO H 0-H JCH3-o-o-CH2-cl: -Br ——§—->25 JCH3_<0-CH2-CH2-Br 2.o6(s)‘ l- 3.50(t) 2.76(s) ' use) . L 3-66 l.\.96(m) (A5 = 0.61 ppm) for the methylene protons attachedito the ester oxygen and the carbonyl methyl group (A5 = 0.70 ppm). Recent work by Olah and 6 coworkers 0 supports these assignments in that carbonyl protonated ethyl acetate was reported to have similar chemical shifts in FSO H-SbF6-SO2 3 {M H + 3 CH c 2.75(s)—/ MHz-L3 ,— 1.h8(t) #9501) as shown.below: 1+0 Figure VI. Reaction of 2-Methowethy1 Acetate with P303}! (Addition of Ester to Acid; Reaction Time - 10 mimites) a _ H H cH3c{CO--(3112-0154)-(3H3(f) mo H WM (1) 0 u + CH -C-O-CH2-Cl12-O-0E3 —-3—> 3 (b)CH<:] Unknown (c), (8): (h) W x 3.10 (mark) r use (n) a 2.71 (a) s 4.72 (s) b 2.71; (s) h 5-05 (m) c 2.78 (a) 1 5.30 (a) d l$.23 (m) e Mu (a) hl Finally an attanpt was made to cyclize allyl acetate with F8033. It was thought that if the olefin's double bond could be protonated in a Markownikoff manner a transient secondary carbonium ion might be generated which could cyclize by participation of the acetoxy group, leading to a 2,’+- dimethyl-l,3-dioxolenim cation. Upon combining these reactants, an exo- thermic reaction ensued leading to a black tar. Nmr analysis of the F8033 decantate revealed only a singlet at 2.77 ppm which was identified as pro- tonated acetic acid. It appears as though carbonyl protonation occurs initially, followed by cleavage to acetic acid and the allyl cation. 0 CH3-g-O-CH2-CH=CHE ——9 CH3F“ +>4CHE+<3 23F ' r' Eé‘CHé' " n ’ ' Hé’ 3:“ r 035012 n o g h + 2AgBr Table IV). Several of the dications, where n = 2, 3 and h, could not be isolated from.the silver bromide due to their insolubility in common polar solvents such as methylene chloride, liquid sulfur dioxide or nitromethane. Interestingly, when n = l, 5 or 6, the dications were isolable as white crystalline products by repeated extractions of the dication-AgBr mixtures with liquid sulfur dioxide. The dications (n=2,3,h) were, however, fairly soluble in FSO3H and could be characterized by extracting the silver bromide— dication.mixtures with this acid. Satisfactory nmr spectra were Obtained by scanning these extracts. (See Spectra h9-5h), Preparation of the dication containing no insulating methylene groups (n=O) between the dioxolenium rings is less straightforward. 'When bis-(2- bromoethyl) oxalate was allowed to react with AgBFh or AngFé, a silver 1&3 TABLE IV 2, 2' -Alkylenebis-l, 3- dioxolenium Dications 2 L42; 1 2C8-210 2 - 3 - u _ 5 173-175 6 198-200 0. 9 - [Qwek slam?“ 262.1% 88 95* 77* 88 91 Reaction Time , hrs . 8.5 1h 3.0 2.5 2.5 2.5 *Yield calculated for a mixture of the dication and AgBr m A9 50 51 52 53 5h uh bromide precipitate appeared after several hours at 25-30°. However, the reaction was not clean and spectral analysis was obscured by a variety of O Br-CHZ-CHQm-g-fi-o-CHQ-cna-Br 2%? [+H+:lg— [:E>-§-O(CH2-)2Br by-products. This was particularly true of the reaction with AgBFh, which produced an intractable paste. The reaction with AngF6 was more promising, but still ambiguous. After a reaction time of 1% hours,a yellow gummy solid 'was Obtained which was characterized.by extracting with FSO H and.examining 3 the resulting extract by nmr spectroscopyu The major resonance signals were a singlet at 5.70 ppm and a multiplet at h.79-5.07 ppm in a ratio of ~2:3. .After a reaction time of 51 hours the reaction product was again analyzed in the same manner. The same signals described above were predominant, accompanied by a weak singlet at 6.00 ppm (see Spectrum.h8). It was inter- esting that the ratio of the singlet at 5.70 ppm and the multiplet at h.79- 5.07 ppm.had changed to:~2.3:l after this longer reaction time. The singlet at 5.70 ppm is certainly appropriate and characteristic for the dioxoleniwm ring protons, but one cannot be certain whether it is due to the monocation or dication. The multiplet is presumably not due to the bis-(2-bromoethyl) oxalate ester precursor which gave an A2X2 pattern at h.87 and 3.6h ppm in FSO3H. Related dications containing no insulating groups between the positive 63,6h centers have recently been postulated by Hoffmann as elusive inter- ‘mediates in the oxidation of tetramethoxyethylene to dimethyl oxalate. #5 O O CH -O\’ O-CH I CH -Q‘ O-CH " | 3 f=< 3 i» 3 tic-C? 3 -v CH3-O-C-(3-O-CH3 CH3-O O-CH3 CH3-O O-CH3 + 2CH3I Bis-(2—bromoethyl) ester precursors were prepared by refluxing the appropriate diacid chlorides with two equivalents of 24bromoethanol in carbon tetrachloride for 6-25.5 hours (see Table V). Yields were 67—96% and the esters were high-boiling liquids or crystallizable solids. These esters exhibited expected nmr (see Spectra 12-18) and infrared spectra as well as giving suitable elemental analyses. Fluorosulfonic acid was an excellent medium for obtaining nmr spectra of these dications. The spectra consisted ofna characteristic sharp sing- let for the equivalent dioxolenium.ring protons as well as appropriate resonance signals for the 2-alkylene groups. In all cases integrations of these signals were consistent with the prOposed dication structures. These dications have been arranged in order of increased shielding of the ring protons in Figure VII. This order should reflect the relative amount of Charge repulsion which results as the number of insulating methyl- ene groups between the electron deficient centers is decreased. As expected, the magnitude of charge repulsion is inversely related to the number of insulating groups. When n = 6 (ion VI) charge repulsion is at a minimum, and the molecule contains two relatively independent, non-interacting cati- onic centers. This can be demonstrated by comparing the ring proton cheme ical shift of the dication (5.32 ppm) with that for the analogous monocat- ion, 2-methyl-l,3-dioxolenium.tetrafluoroborate, in which the ring protons have a chemical shift of 5.30 ppm" ‘When n = 1 (ion I) charge repulsion is at a maximum. This is reflected not only by the large deshielding ID #6 TABLE V Bis-2-(Bromoethyl) Esters o o Br-CH2-CH2-o-g-(CH2)n-5-o-CH2-CH2-Br Reaction Nmr Mp or Bp, ° Yield Timethrs. Spectrum Mp 5h.5-56 96 9 12 l25-6/O.5mm 7h 6 13 1u5-1u7/2mm 67 in in 115-118/2.5mm 7h 12 15 l62-3/O.9mm 79 12 16 156-7/2mm 67 8 17 Mp 37-38 9h - 25.5 18 1‘7 Figure VII 2, 2' -Alkylenebis-l, 3- dioxolenium Dications in order of Increased Shielding of Ring Protons Exchange f'” with FSO.H r—‘ 3.70 (S) 3 . , 0. . ['E)>w2-<+ I 5.61 (s) | ?CH2-Cfle<+ I 5A8 (S) I II 2.16 (q) 2-02 (m) /_ 3.26 (t) /O_ 3.10 (m) DORE-CH2 -CH 2—<’:] 5.1+o (s) [:>-CHE-IV 0112- 032- -CH2<+ :3 5.36 (5) III . 2.36-1.36 (m) / 3.06 (m) 0‘ / ,0 [-3>CH2-CH2-CH2-CH2—CH2-<{;] 5131+ (S) 2.20-1.35 (m) / /——3.os m s‘ r A ‘ 'f l +.'>CH2'CH2'CH2‘CH2‘CH2'CH2'§O:I 5'32 (8) VI us of the ring protons but also by the fact that the insulating methylene protons are so acidic that they exchange readily with theFSOBH solvent. The exchange probably occurs via the equilibrium shown below. It is £20243] 2 E §>r<€3 interesting that the methylene group in a l, 3-dihydroxydicarbonium ion re- ported recently by Brouwer59 does not exchange very rapidly at -20° to +30° in an HF - SbF medium. The nmr spectrum of this dication consists 5 o 0 OH OH CH g CH 5 CH HF-SbF5 5- CH C CH C CH 3'2"3 ’ 3'9, 2'6 3 of three singlets in a ratio of 1:1:3 which were assigned to the 0H, CH2, and CH3 groups, respectively. Recent nmr studies by Dewar65 and Frankel66 have reapened the ques- tion concerning the roles of inductive and field effects in the trans- mission of charge in a molecule. These effects have been previously de- fined by Roberts.67 The smooth monotonic curve that is obtained by plot- ting dioxolenium ring proton chemical shifts as a function of charge separation (n), (seeTFigure VIII) suggests that inductive effects predomin- ate in this system. Successive polarization of 0 bonds, as described by Branch and Calvin, 68 accounts for the observed charge repulsions. The charge attenuation per bond in this system appears to be of approximately the same order as that described by these workers, i.e. 2—15' Assuming this value, one would expect charge repulsion to nearly disappear when n = 5-6. n = l 2 3 1+ 5 6 wag—CHE... egg—m ....2< 5;] } ...... Chg = 0. 36 0.13 .0h5 .016 .006 .002 1+9 5.70 5.50 I- 0 5.50 - 8 5.40 - 5.30 *- 530 I I I l I l O l 2 3 4 5 6 7 nouns m DIOXOLENIUM RING PROTON CHEMICAL SHIFT vs NUMBER OF METHYLENE GROUPS 3.70 f 3.60- 3.50— 3.40- 8 3.30— 3.20- 5.10— 300 1 1 1 1 1 n o 1 2 3 4 5 6 7 FIGURE 1x; Z'METHYLENE PROTON CHEMICAL SHIFT VS NUMBER OF METHYLENE GROUPS 50 This is indeed the case; dication VI, where n = 6, has approximately the same ring proton deshielding value as the monocation, 2-methyl-l,3-dioxol- enium tetrafluoroborate. Further examination of this system showed that these data are fit rather well by the empirically derived equation (1) where 5.30, the limit- ing value of 5 when n = 0°, is identical with the chemical shift of the 5 = 5.30 {1'602 (1) (n+1) ring protons in the 2-methyl-l, 3-dioxolenium monocation. When n is large, the molecule behaves as if there were no interaction between the cationic centers. Equation (1) obviously fails when n = 0, but attempts to fit the data with an equation using n + l, n2 + 1, etc. in the denominator of the last term, with appropriate changes in the parameters, were unsuccessful. The chemical shifts of the methylene protons between the cationic centers provide an even more sensitive probe of charge distribution in these dications. Figure DC shows a plot of the a-methylene proton chemi- cal shifts as a function of n; these data are fit almost precisely by the equation. In Figures VIII and IX, the solid curves are drawn accord- 5 = 2.91. (1 ”-3.1 (2) n ing to the equations, whereas the points are experimental. Equation (2) suggests that the chemical shift of the a-methylene pro- tons can be described as the sum of two terms, one of which is constant (i.e., the methylene is always adjacent to and a constant distance from one dioxolenium ring) and one of which varies inversely as the square of n, which may be proportional to the distance from the second cationic center. An alternative, very simple way of correlating these data is also possible. Each methylene in ion II may be considered to be 0: to one 51 dioxolenium.ring and B to the other. The chemical shift can be expressed as the sum of two parameters, 6a:+ 0 Values of these empirical constants B. which fit the data are: 60:: 2.h7, 6B Y 0.59. The data of Figure VII provide only one independent check on the = 1.23, 6 = 0.79, 56 = 0.63, be = method, which happens to give a calculated 6 that agrees exactly with the observed. However, this method has also been applied to analogous data reported by Olah and coworkers for bis-acylium.ions.138 The best para- meters are 6a,: 3.60, 65 = 1.67, oY = 0.87, 66 = 0.69, be = 0.53, 6: = 0.43, on = O.hl. Two examples of the closeness of fit are <3 (n.29) (2.5u) e> Ch==il c C c c C h.20 2.52 and e (11.01) (2.10) (1.110) (1.38) (p 0===0—————<>—————C " C————-—C——-——«: ——c -C n.00 2.10 1.h0 1.29 where values in.parentheses are calculated, and those below the formulas l were observed by Olah and Comisarow.:¥The parameters for bis-acylium ions are larger than those for bis-dioxoleniwm ions, due to less charge delocal- ization in the former. C. 2-Aryl-l,3¢dioxolenium Cations: Using a modified version of Meerweinfs method,22 fifteen.2-(mg§a_and page substituted aryl)-l,3-dioxolenium cations were prepared by reacting equimolar amounts of appropriate 2-bromoethyl esters with anhydrous silver tetrafluordborate in.methylene chloride. Stirring these reactants for l—2 hours at 25-30° gave 55-96% yields of the white tetrafluoroborate salts (see Table VI). Precautions Observed in the workup and storage of these cations were essentially the same as described for the 2-alkyl-l,3-dioxol- enium cations. Separation of these cations from silver bromide was Agl Substituent prethoxy 3,8,5-Trimethoxy prethyl mrMethyl Phenyl pfFluoro prhloro m— Chloro mrBromo z_n_-; Fluoro 3,8-Dichloro r_n_- ‘I'r ifluoromethyl p- Tr ifluoromethyl mrNitro peNitro 52 TABLE VI 2-Aryl-l,3-Dioxolenium.Cations IO '- Ary14<+ I x ‘0 M 228-230 166-167.5 207-209 l9h-l96 168-170 203-205 235-237 173-175 178.5-176 169-171 218-220 135-137 188-189.5 lh9-lSl.5 209-211 Yield 96 77 81 76 81 63 82 69 61 61 69 72 63 7O 55 Reaction 22852.23; Nmr Spectrum 55 56 57 58 59 60 61 62 63 6h 65 66 67 68 69 52 TABLE VI 2-Aryl- l, 3-Dioxolenium Cations O - Aryl , tin 43.3.6. .68.: oz... 3223963 .8 58 D. 2,2T and 2,2:L¥"-Aryl-l,3:Dioxolenium.Dications and Trications When di- and trisubstituted 2—bromoethyl aryl esters were treated with equivalent amounts of silver tetrafluordborate in the usual manner, several interesting l,3-dioxolenium dications and a trication'were obtained. 2-Bromoethyl isophthalate and 2—bromoethyl terephthalate gave the corres- ponding meta and para dications, respectively. Similarly, 2-bromoethyl O Br-CH -CH -0-C -O-CH -CH -Br '——-—-4’ {" 2BF 2 2 2 2 Ix CH2C12 O 0 II n MAEBF Br-CHQ-CHe-O-C- -C-O-CH2-CH2-Brfi EBFh CH2 Cl2 trimesate led to the symmetrical trication. The multications were insoluble if C-O-CH —CH -Br 0 2 2 . Br-CH -CH -O-C. --——€> 33F 2 2 - h CH2Cl2 g-O-CHe-CH2-Br in methylene chloride, liquid sulfur dioxide and thionyl chloride and could be isolated only as mixtures with silver bromide. Identification of these cations was possible, however, by extracting the AgBr-multication.mixtures With FSO3H and filtering through a sintered glass funnel. Satisfactory nmr Spectra were obtained by scanning these filtrates. The characteristic A2X2 Pattern observed for the ester precursors (see Spectra 3h—36) were absent in the spectra of the ions, which contained only highly deshielded singlets for the equivalent dioxolenium.ring protons accompanied by appropriate 59 resonance signals for the aromatic nuclei (see Spectra 70-72). Chemical shifts of the dioxolenium.ring protons were in the expected order as shown in Table X. The electron deficiency of the polycations is reflected not only by the highly deshielded protons of the dioxoleniwm rings, but also by the low field positions of the aromatic protons. TABLE X Nmr Chemical Shifts for 2-Aryl-l,3-Dioxolenium Multications in FSO3H l or 2 Dioxolenium.» Cation Ring Protons Aromatic Protons 5.60 (s) 8.03 to 9.26 (m) 0 E“ © a+ 2x“ 5.63 (s) 8.72 (s) 3x‘ 5.77 (s) 9.58 (s) 60 These multications serve as excellent models for demonstrating the use of l,3-dioxolenium.ring protons as electron density probes to determine the Hammett O values of hydrolytically unstable moieties. For example, in the mgtgfsubstituted dication one can assume that one of the dioxolenium rings is the probe whereas the other ring is the mgthsubstituent. By extrapolating the Hammett plot for the fifteen I_n_e_t£ and pa_r_a_ substituted cations (Figure x) until it intersects with the nmr chemical shift for the mgtgydication one can Obtain a a value for the l,3-dioxolenium.moiety as a mgtg_substituent. The 0 value was found to be +0.8h. A similar treat- ment of the pargfsubstituted dication gave a 0 value for the l,3-dioxo- lenium.moiety,as a page Substituent of +0.97. This latter value is the .largest positive p§£§_ 0 value thus far reported. The symmetrical trication provided an excellent model for checking both the accuracy of the determined me_ta Hamett 0 value for the l, 3- dioxolenium.ring and the reliability of the linear plot of the fifteen monocatflzs as a function of their 0 Substituent constants. UBing the additivity principle described by Jaffe69 for multiple m§t§_or EEEE'Sub- stituted benzenes, it was predicted from the determined metg_ 0 value (2 x 0.8h = +1.68) that the trication dioxolenium ring protons should have a Chemical shift (6) of 5.80 ppm, This value was Obtained by intersecting the extrapolated Hammett correlation plot with a line perpendicular to the abscissa at o = +1.68. The observed chemical shift for the heterocyclic ring protons of the trication was 5.77 ppm and is designated V on the plot in Figure X. This is in excellent agreement with the predicted value and soundly corrOborates the entire correlation. IV DISCUSS ION 62 The four families of cations which we have examined in the l,3-dioxo- lenium ion series can'be divided into distinct and distinguishable groups according to the chemical shifts of the ring protons (Figure XI). The chem- ical shift gradient varies from the most shielded member of the 2-alkyl- l,3-dioxolenium family (i.e. 2-diethylamino-l,3-dioxolenium ion) where o = h.98 ppm to the most deshielded member of the aryl multication family (i.e. 2,2',2"-phenylenetris-l,3-dioxolenium.trication) where 5 = 5.77 ppm. Ignoring anisotropy effects of the 2-substituents, this gradient appears to represent, to the first approximation, the relative order of ground state energies of these alkoxycarboniwm ions. In accordance with this hypothesis, the relative order of members of families as well as the order of the families can'be rationalized in terms of resonance or inductive interaction of the cation with its 2-substituents. These aspects will be considered in detail according to family. 2-Alhyl—lL3-Dioxolenium.Cations: As shown in Figure XI the 2—alkyl-l,3-dioxolenium.cation family is the most shielded group in the series. If one examines the chemical shifts for members of this fami y (Figure XII) it is apparent that cation I is a special case and should perhaps be thought of as a member of still another cation family which consists of cations containing heteroatoms as 2-substit- uents. Cation I is presumably the most stabilized member of this entire series due to its ability to delocalize charge from the ring to the nitrogen atom and is perhaps best represented as an immonium ion. Et , Et\+ Et\+ \N :+ :::] <+——e> ‘::] Et/ ‘ _ Et/ : Et’ C 63 Figure XI «20.81.. 02;. uzhmo mph—Em 43.298 .m .3 .2 m> _ $3.25. 5:8 3223983-n._-oH.=E:mm=m-~ 8v cod FIGURE III _<9 Alel it] . O u .I2 a 9 IO ' 5.25 _ 4 5 o ‘ g o ’ I (EH2N 11A;- m (maize-cu- 5.oo- ' 1v cns-o-Q-cn-cu- ' v CH3CH-CH- v1 ©— CH -¢H- VII H0- 4.75 — VIII X CH3 " 2H3 x1 CH2- m CHz'CH- 4.50 65 It is not known whether one can as a rule expect this kind of stabil- ization in analogous systems containing oxygen or sulfur as a 2-substituent. In the case of 2—hydroxy substituted cation (VII), relatively little delo- calization of charge from the ring was noted compared to the 2—alkyl sub- stituted cations. Cation (VII) was more deshielded than five of the ten 2-alkyl substituted cations examined in Figure XII. This suggests that the hydroxy oxygen atom contributes very little to delocalization of charge and parallels similar Observations made by Taft and Ramsey21 on analogous acyclic systems. This anomaly may be due to protonation of the ether Hess: <—» so <—-> s2] oxygens of the ester whereby charge delocalization would not be possible. 60 70 This mode of ester protonation has not yet been observed. ’ 2—Cyclopropyl-l,3-dioxoleniwm ion (II), Figure XII, was the most shielded example of this family where the Substituent was entirely hydro- carbon. Similarly in Table XI it can'be seen that both thetz and B-hydro- gens of the cyclopropyl groups are deshielded substantially. This demon- strates a large amount of delocalization of charge into the cyclopropane ring. This enhanced delocalization is probably due to interaction of the cyclopropane "bent bonds” with the vacant prorbital of the cyclic dialkoxy carbonium ion as shown below. This kind of interaction has been invoked by 66 HartYl, Olah72 and Den073 to account for the unusual stability of other related cyclopropyl carbonium ion systems. subsequent work by Schleyerrfl4 and Richey75 has provided further evidence for the nature of this interac- tion, which is thought to be best represented by the bisected conformation shown above. It is interesting that the other small ring system in this family, 2—cyclobutyl cation (VIII), was not stabilized much more than the 2-methyl or 2—tfbutyl cations, X and IX respectively. Similarly, Table XI shows that relatively little delocalization of charge to the cycldbutane ring has occurred since the amend B protons are deshielded only slightly. It is generally accepted that in hydrocarbon systems, allylic and methylallylic carbonium ions possess enhanced stability compared to satur- 12f,76 ated systems because of their ability to delocalize charge. This delocalization has been.prOposed by Simonetta and Heil‘bronner77 to occur R R CH I ' 2§\ 0¢0H2 <-—-> CH2=C-CH2® 4—) ale-R - CH2 by means of l,3-fi interactions between the terminal centers. Valence bond treatment in this manner gave the following charge distribution. Hirst of ca 2 (0.367 (0.367) and Linnett78 preferred to invoke the tradional resonance argument and by a Simple Hflckel treatment of the two odd alternant terminal allyl cations, Suggest that the positive charge should reside predominately on the term- inal carbon atoms. Large deshielding of the methine proton (A5 = h.12 ppm) and the methyl group (A6 = 2.38 ppm) was observed in the allyl and 66a .- TABLE XI Comparison of Proton Chemical Shifts of 2—Substituents in Ester Precursors (CCl ) with those in 2-Alkyl- l, 3-dioxolenium Cations (FSO3H) 2-Substituent Ester Cation I (Et)2-N- a. l.l3(t) a. 1.32 b. 3.27(q) b. 3.59 II a. 0 73-l.20 a. 1.75-2.07 m) [>— b. l h0-l.89( b. 2.10-2.53ém) III (CH3)20=CH- a. 2.03(d) a. 2.37éd) b. 5.66 (m b. 6.27m) IV CH 0 H=CH- a. 3.80(s) a. h.02(s) 3 © b. 6.23 d) b. * ' c. 6.83 d) c. * d. 7.h3 d) d. * e. 7.60 d) e. * v CH3-CH=CH- a. l.92(d) a. 2.27 d) b. 5.823s) b. 6.1m d) c. 7.01 m) c. 8.22 m) VI '—\-CH=CH- a. 6. 37 d) a. 6.86 6.) Q b. 7. he m) b- 7-&§m) c. 7. 66(d) c. 8.59 d) VIII (\>- a. 2.1hém) a. 2.35?) V b. 3.12 q) b. 3.66 q) 13: (CH3)3C- a. l.20(s) a. 1.50 x CH3- 3. 2.06(s) a. 2.75 $33 XI CH2=C- a. 1.96és) a. 2.15(s) b. 5.87 a) b. 6.85(d) XII CH2=CH- a. 5.68-5.99(m) a. * b. 6.12-6.65(m) b. * *SPe'ctrum too complex to analyze 67 methylallyl cations, respectively. This led Olah12f to suggest that l, 3.11 interactions contribute strongly to the delocalization of charge. Close examination of related allylic carbonium ions III, V, VI and XI in the dioxolenium series reveals the following trends; hydrogens on the double bond 0: to the positive center are deshielded from their ester (A6=0.6l) (A6=O.35)CH3\ ,H (A6=O.62) OH H 3\ A = o 1" = = ( 5 ° 3 )cs/ (fig (A6=l.21)H; CF 3 III v (A0=O.22) (A6=0.)+9) H H (A6=O.19) =c’H (A6=0.98) )=v$;3 (A6=0.93) H VI XI precursors A6=0.h9-0.62 ppm (e.g. III,V,VI). The a methyl group, in XI, was deshielded somewhat less, A6=O.l9 ppm. It is especially noteworthy that protons gig or 39.1.1.9. B to the positive center are highly deshielded and are shifted from 0.93-1.21 ppm downfield from their esters (e.g. see V, VI and X1). These protons are deshielded approximately twice as much as their a counterparts. Similarly gig and Eggs; B methyl groups (e.g. III and V) are deshielded 0.3h-0.35 ppm fran their esters. Again this is about twice as much as is observed for an a methyl group (e.g. XI). If one as- sumes these deshielding values reflect charge densities and are not due to anisotropy effects, it is tempting to compare these values to the charge densities calculated by Simonetta and Heilbronner'r7 for the l, 3-TT inter- action model. From this model it is predicted, according to the following calculation, that approximately 73% of the charge resides on the 68 % Positive Charge on (2)(O.367) 2 13% Terminal Carbons in = 10 = l,3-n Interaction.Model (2)(0-367)+(0-266) terminal carbons and 27% on the internal carbon. Using deshielding values (A0) for thecu and B-proton probes in Cation V it is calculated, as shown below, that approximately 80% of the charge is on the terminal carbon and a~20% on the internal carbon. Similar treatment of the values fort: and B % Positive Charge on 2(A5 for 8-H) 2 Terminal Carbons = 10 = in Cation V 2(A0 for B’fi)+(A5 for a-H) ' (2m. 21) 2 _ (2)(I.2l)+(0.62) 10 - 22% protons in cation VI predicts the same values. UBing the A6 value for the % Positive Charge on 2 O. 2 Terminal Carbons in - 10 = 80% Cation VI 2 0'93 + 0' 9 a-methyl group in cation XI and the A6 value for the B_methyl groups in either cation III or V one obtains almost the same value for charge delo- calization in these cations. A comparison of these crude delocalization % Positive Charge on 2 0 2 Terminal Carbons in = 2 0 35 19 10 = 72% Cations III, V and XI ° ' data to the value obtained from the Simonetta-Heilbronner model provides supporting evidence for such l,3-n interactions in the vinyl, propenyl, isopropenyl and styryl—l,3-dioxolenium cations. This interaction might be represented as follows: In view of the possibility of this kind of delocalization it was rather surprising to find that the dioxolenium ring protons in the 2-vinvl and 2-isopropenyl-l,3-dioxolenium cations were deshielded from saturated systems such as 2-methyl and 2—t-butyl cations (see Figure XII). However, when the unsaturated substituent contains a terminal group which is capable of stabilizing positive charge, the cations such as III, IV, V and VI appear to resume their predicted ability to delocalize charge from the dioxolenium ring. According to the deshielding of the dioxolenium ring protons the unsaturated cations fall in the following order: (CH3)2-C=CH- < p-CH30-¢-CH=CH- < CH3-CH=CH- < ¢-CH=CH- < III IV V VI CH3 I Saturated Cations VIII, IX, x (032:0- < cafes- XI XII In order to rationalize the relative order of these cations one must con- sider the dual oxonimn-carbormm ion character of these systems. The anom- alous order of the two unsaturated cations XI and XII may reflect the high order of their oxonium ion character (b); such structures would be stabil- 0 + R-<+j H Réj 0 (a) (b) ized by conjugation with the carbon-carbon double bond. The higher 70 electronegat ivity of the sp2 hydridized carbon substituent and the predom- inate oxonium ion character of XI and XII may account for the fact that the ring protons in these cations are more deshielded than in the saturated cations VIII-X. In the case of the terminal substituted unsaturated cations III, IV, V and VI more interaction of the double bond with the positive charge occurs since the cation resulting from allylic type resonance can be stabilized by methyl or phenyl groups. , o 0 .3_ R-CH=CH-€j <—-> R—CH-CH CHBC‘CH-CH=<:j 2L2”-Alkylenebis—l,3-Dioxolenium.Dications: Ring proton chemical shifts for members of this family where n = 5 or 6 (5 = 5.32-5.3h) do not differ much from the saturated members of the mono- catitm1(6 = 5.30) family. This clearly indicates that these dications are reLieved of charge repulsion interactions and probably have ground state energies which are quite similar to the monocations. Decreasing the number of insulating groups between the positive centers dramatically increases dufltge repulsion, especially when n = l or 2. This is well illustrated in 72 Figures VIII and IX. In fact when n = l the charge repulsion is so great that the a protons are acidic and exchange readily in acid solution. Ex- change by means of the monocation shown below provides relief from these [5) 0%] ii. {13* repulsion interactions. Although exchange of a protons was not observed for n = 2—6 in fluorosulfonic acid, the acidity of these protons probably increases as n decreases from 6 4 2. This may account for the unusual sensitivity that the a protons display when utilized as a probe (see Figure IX). This dication system provides a unique model for examining the effects of charge repulsion as a function of separation and nmr spectroscopy pro- vides an excellent tool for assessing these parameters. Using the empiri- cally derived equation (2), presented in the Results section, it is possible 5 = 2.9h (l + -%5) n to obtain an excellent correlation of a-proton chemical shifts with charge separation both in our system as well as Olah's bisacylium dication ser- 13a ies. Examination of this equation reveals its similarity to the well known Coulomb law79 which defines the force between two charges as follows: (Charge)@hameL = 9:91. Force = (nielectric Constant)(distance)2 Kn? By simple mathematical operations on these two equations it is possible to derive an equation which will allow the calculation of repulsion forces in these dication systems by merely knowing the chemical shift of the a-protons 73 and the dielectric constant of the medium separating the charges. The derivation is as follows: Dividing empirical equation by 2.9h gives: 6 l = l + —- (l) 2.91; n2 Solving for -$2-, one obtains: n l 6 32‘ = 5751: - 1 ‘2’ The Coulomb equation is reduced to a more simple form with the following assumptions: The dication charges are equal, therefore: 2 F = 9‘5 (3) Kn Multiplying equation (3) by 492-: Q 3‘5 = i; (a) Q n Equating 3'5 in equations (2) and (h) one obtains: n 35"5735'1 ‘5’ Q Solving for F one obtains an equation for calculating the repulsion forces in these dications which only requires knowledge of the chanical shift of the a-protons, magnitude of the charge and the dielectric constant of the medium separating the charges. The equation is as follows: Repulsion 3 Q3 6 _ ‘ [Bree K (2.95 1)“) This treatment should be applicable to other dicarbonium ion series such as l3a,80 those reported by Olah and coworkers. Such an examination could 7h provide further insight as to the relative ground state energies of these dicarbonium ion systems. 2-Apyl-l,3-Dioxoleniwm Cations: As shown in the Results section, 2-(papa_and papa substituted) aryl- l,3-dioxolenium.ions exhibit a very good linear free energy relationship in their nmr correlation with Hammett 0 values. In view of the possible carbonium ion character of the dioxolenium ring, it was at first surprising that Brown o+' values81 gave a much poorer correlation. However, hydrol- ysis studies on pfsubstituted methyl orthobenzoates also show a better cor- relation with Hammett 0 values than with 0+ values.81 These reactions have since been shown to involve related aqwflic dialkoxy carbonium ions. These observations suggest that resonance interactions between electron- supplying Substituents and the electron deficient l,3-dioxolenium moiety are not strong. This may indicate that the oxonium.form.(c) is an important + ,0— 0 0 a, <—» Bo «—> Bo (10 ) (a) C) contributor to the resonance hybrid (a). Nevertheless, delocalization of charge to the EEEE position of the phenyl ring is apparent if one examines the relative amounts of deshielding <3f the papa and papaymethoxy groups in the 3,h,5-trimethoxy phenyl-l,3- cm: the positive charges are on adjacent carbons. From these classical reso- O nance representations it is apparent that, based on charge interaction, III should have a higher ground state energy than II. Similar delocalization of charge in the trication, III, can lead to only one contributor as shown 78 below: The greater accumulated formal charge in the trication undoubtedly gives this cation the highest ground state energy of the entire series. As shown in Figure XIII the deshielding parameters (A6) for the aro- matic nuclei in each case parallel the relative ground state energies pre- dicted for the members of this family and are consistent with the conjec- tured delocalization and charge interactions described above. 79 FIGURE XIII I Deshielding Parameters (A6) for Mono-, Di- and Tri-Dioxolenium Cations (FSO3H) Compared to their Ester Precursors (CClh) A6=O.3O {HS 6:5.h2 ppm ppm ‘ Emil .. L_,r_n ‘ A6=O.57 ppm III s40 f)" 5 Q9 II 3:) 5=5.77 ppm 0 “\K—vA6=l.0h ppm V . EXPERD’IENTAL 81 A. General 1. Melting Points: .Melting points were measured on a Thomas-Hoover melting point apparatus and are uncorrected unless otherwise specified. 2. Microanalyses: Elemental analyses were carried out by MI. L. E. Swim, Analytical Laboratories, The Dow Chemical Company, Midland, Michigan. 3. Nuclear Magnetic Resonance Spectra: Spectra were scanned on a Varian Model A-6O instrument using either tetramethylsilane or tetra- methylammonium.tetrafluoroborate, as specified, for an internal standard. The chemical shift for tetramethylammonium.tetrafluoroborate was taken as 3.10 ppm. .All chemical shifts are reported as 6 values in.ppm. 4. Infrared Spectra: Infrared spectra were recorded on a Perkin- Elmer Model 337 spectrometer either on NaCl or KBr plates, as indicated. 5. Solvents: Anhydrous methylene chloride and acetonitrile were prepared by distilling these solvents from phosphorous pentoxide; they were then stored over silica gel (Grace Chemical Company). Anhydrous diethyl ether (Mallindnxflt) was used as it was received without further treatment. 6. Miscellaneous: All recrystallizations and.manipulations of the moisture sensitive cations were carried out in a Labconco dry box contain- ing several (6 x 12 int) beds of phosphorous pentoxide. Bottled cation samples were always stored in a desiccator over anhydrous calcium chloride. B. .Precursors to 2-Alky1-l,3-dioxolenium Cations l. 2-Bromoethyl Acetate: This material was obtained from the East- man Kodak Company. Redistillation.through a 1/2 x 21"'Vigreux column gave a colorless liquid, bp l60-l6l° (Nmr: Spectrum.9). 2. 2-Bromoethyl Acrylate: This compound was obtained from the Bordon Chemical Company as a tan liquid. Using N,N-diphenyl phenylene 82 diamine (DPPD) as an inhibitor, both in the distillation pot and receiver, this product was redistilled through a 1/2 x 21" Vigreux column to give a colorless liquid, bp 52—53°/5 mm.(Nmr: Spectrum 11). 3. 2—Bromoethyl Methacrylate: This was Obtained from the Borden Chemical Company. This material was redistilled (inhibited with DPPD) through a 1/2 x 21" Vigreux column to give a colorless liquid, bp h6-h9°/ 2.7 mm. (Nmr: Spectrum 10). h. 2-Methoxyetpyl Acetate: This material was obtained from the Eastman Kodak Company. Redistillation through a 1/2 x 21" Vigreux column yielded a colorless, sweet smelling liquid boiling at lhl-lh3°. 5. 2-Hydroxyethyl Acetate: This material was Obtained from the Eastman Kodak Company. It was redistilled through a 1/2 x 12" Vigreux column. A colorless fraction'boiling at 185-188° was collected. 6. 1,2-Diacetoxyethane: This material was obtained from the East- man.Kodak Company and was redistilled through a 1/2 x 12" Vigreux column. .A colorless fraction'boiling at 190-192° was collected. 7. Preparation of 2—Bromoethyl—N,N-Diethyl Carbamic Acid Ester: N, N-Diethylcarbamoyl chloride (Eastman, 27.12 g, 0.2 mole) and 2-bromo- ethanol (25 g, 0.2 mole) were combined with 50 ml of carbon tetrachloride aini refluxed for 20 hours. Solvent was removed from the dark'brown reac- ixhon.mixture with a rotating evaporator (Buchi). The amber liquid resi- g=0’ 1728 cm-1 (Fluorolube). gap: Spectrum 5. 13. Preparation of 2-Bromoethy1 Crotonate: 2—Bromoethanol (h0.25 g, 0.5 mole) was added over a period of 5-10 minutes to 52 g (0.5 mole) of crotonyl chloride which was being stirred in a 250 ml three-necked flask equipped with a condenser, stirrer and addition funnel. After the addition was complete, an exothermic reaction set in and was accompanied ‘by the evolution of c0pious amounts of hydrogen chloride. The ten reac- tion.mixture was maintained at reflux, while stirring for h hours. During that time the mixture became dark'brown. The crude product was distilled through a 1/2 x 21" Vigreux column giving a colorless, major fraction boiling at 110-117°/39 mm which weighed 19.5 g. This material was redis- tilled to give a colorless, major fraction boiling at 118-119°/40 mm. Apal: Calculated for 06H9Br02: C, 37.3; H, h.70. Found: C, 37-3; H, 1857- Ipfpapaa: "g=0’ 1735 cmfl;i\g=C/-, 1668 cm.-1 (neat). I / \ yap: Spectrum A. 14. Preparation of 2—Bromoethy1 Cinnamate: Cinnamoyl chloride (Eastman, h9.9 g, 0.3 mole) and 2—bromoethanol were combined with 75 m1 of carbon tetrachloride in a 250 m1 roundebottomed flask equipped with a reflux condenser. The reaction mixture was refluxed for 2h hours and 86 the solvent was removed on a Bfichi rotating evaporator to give a viscous, tan liquid residue. Distillation of this material through a 1/2 x 12" Vigreux column yielded a major fraction boiling at 120-123°/2 mm which was very prone to solidify in the condenser and receiver. The crude, white solid product (h9.5 g, 62%) melted at hl-h5° and was very soluble in diethyl ether, acetone and carbon tetrachloride. Recrystallization from diethyl ether using a Dry loam—methylene chloride cooling bath pro- duced nice white needles melting at h3-h6°. The melting point was imr proved to hh.5-h6° by placing the product in a vacuum desiccator over paraffin shavings and evacuating to 2 mm. apal: Calculated for CllHlIBrO2: C, 51.8; H, b.35. Found: C, 52.1; H, b..63. Infrared: :g=0’ 1719 cm.1 (Fluorolube). flap: Spectrum 6. 15. Preparation of 2-Bromoethyl Pivalate: Pivaloyl chloride (Eastman, 36.18 g, 0.3 mole) and 2-bromoethanol (37.5 g, 0.3 mole) were combined with 50 ml of carbon tetrachloride in a 250 ml roundabottomed flask equipped with a reflux condenser. The reaction.mixture was re- fluxed for 10 hours after which the solvent was removed on a Buchi rotat- ing evaporator to yield a light yellow oil residue. This crude product 'was distilled through a 1/2 x 12" Vigreux column to give a fraction boil- ing at 75-85°/33 mm which was found by nmr spectroscopy to be predomin- ately the desired product. Redistillation through a 1/2 x 21" Vigreux column gave 15.6 g (25%) of colorless product boiling at 92+-95°/3u mm. Anal: Calculated for C7Hl3BrO2: C, no.2; H, 6.27. Found: C, C, no.0; H, 6.18. Infrared: \7 , 1738 cm.1 (neat). —— ,C=0 Nmr: Spectrum 8 87 16. Preparation of 2-Bromoethy1 CyclObutanecarboxylate: A solution of cyclobutanecarbonyl chloride (Kaplop labs, 26.9 g, 0.2 mole) in 25 m1 of carbon tetrachloride was charged into a 250 ml three-necked flask equipped with a stirrer, condenser, and addition funnel. A solution of 2-bromoethanol (Eastman) (25 g, 0.2 mole) in 25 ml of carbon tetrachloride was added dropwise with stirring over a period of 20 minutes. The reac- tion mixture was stirred while refluxing for 15 hours. Removal of the solvent on a Bfichi rotating evaporator gave a sweet-smelling liquid resi- due which was distilled through a 1/2 x 21" Vigreux column. A water white product boiling at 79-80°/5 mm was obtained (31.3 g, 77%). gag: Calculated for C7HllBr02: C, 1+0.6; H, 5.36. Found: C, 110.14; H, 5.27. Infrared: 7 l7)+0 cm”l (neat) _— >Cgo’ m: Spectrum 7 C. Precursors to 2-Ayl-1L3-dioxolenium Cations: 1. Preparation of 2-Bromoethylp-Methombenzoatex p-Methoxybenzoyl chloride (Aldrich, 51.2 g, 0. 3 mole) and 2-bromoethanol (37. 5 g, 0. 3 mole) were combined in 50 ml of carbon tetrachloride and refluxed for 15 hours. When the mixture was cooled to room temperature, some white crystalline material fell out of solution and was filtered off and identified as h- methoxybenzoic acid. Solvent was removed from the syrupy filtrate on a rotating evaporator leaving a dark brown, viscous residue which was dis- tilled through a 1/2 x 21" Vigreux column. A major fraction came over at 109-110°/3 mm (39.1 g, hut). Anal: Calculated for ClOHllBr03: C, 1+6.h; H, h.28. Found: C, 1+6.l+; H, I+.35. Infrared: i=0, 1722 cm"1 (neat) Nmr: Spectrum 19 88 2. Preparation of 2-Bromoethyl 3,&,5-Trimethoxybenzoate: 3,h,5- trimethoxybenzoyl chloride (Aldrich, 69.2 g, 0.3 mole) and 2-bromoethanol (37.5 g, 0.3 mole) were combined in 50 m1 of carbon tetrachloride and the orange-red reaction mixture was refluxed for 10 hours. Solvent was re- moved on a rotating evaporator to give an amber oil that slowly solidified into an off-white mass. Recrystallization of this crude material from n—hexane gave a white fluffy material melting at 565-59" (75.95 $791.). Recrystallization from diethyl ether gave white crystals melting at 56-57°. Anal: Calculated for C12H1505Br: C, h5.2; H, h.7h. Found: C, h5.5; H, n.93. Infrared; :gzo, 1715 cm.1 (Fluorolube) Nmr: Spectrum.20 3. Preparation of 2-Bromoethylprethylbenzoate: prethylbenzoyl chloride (Aldrich, 35.7 g, 0.2 mole) and 2-bromoethanol (25.0 g, 0.2 mole) were combined with no ml of carbon tetrachloride and refluxed for 12 hours. The solvent was removed on a rotating evaporator leaving an amber colored liquid residue. Fractionation of this residue through a 1/2 x 21" Vigreux column gave a colorless major cut boiling at ll6-ll7°/h mm (35.1 s, 71%). apap: Calculated for ClOHllBr02: c, h9.h; H, n.56. Found: C, h9.6; H, h.69. Infrared: 1725 cm-1 (neat) \7 ’ ,c=0 Nmr: Spectrum 21 89 4. Preparation of 2—Bromoethyl kMethylbenzoate: p—Methylbenzoyl chloride (Research Organic Company, 31.1 g, 0.2 mole) and 2-bromoethanol (25 g, 0.2 mole) were combined in 30 ml of carbon tetrachloride and re- fluxed for 12 hours. RemoVal of the solvent on a rotating evaporator and distillation of the liquid residue gave a colorless major fraction'boil- ing at 115-116°/2 m (39.85 g, 81%). Apap: Calculated for ClOHIlBrO2: C, 49.4; H, 4.56. Found: C, 49.3; H, 4.44. Infrared: >34), 1730 cm'1 (neat) Nmr: Spectrum 22 5. Premration of 2-Bromoethyl Benzoate: (a) Benzoyl chloride (Heyden Chemical Company, 28.2 g, 0.2 mole) and 2-bromoethanol (25 g, 0.2 mole) were combined in 50 ml of carbon tet- rachloride contained in a 250 m1 three-necked flask equipped with a con- denser and stirrer. The reaction mixture was maintained at reflux while stirring for 19 hours. Solvent was removed on a rotating evaporator (Buchi) yielding a light amber oil which weighed 44.5 g (97%). This crude product was distilled through a 1/2 x 21" Vigreux column giving a major cut of colorless product boiling at 96-97°/2 mm which weighed 31.4 g (69%). A_napl_: Calculated for C9H9Br02: C, 47.2; H, 3.96. Found: C, 47.2; H, 1+.o9. :0, 1731 cm'1 (neat). _1\§_nr_: Spectrum 23 Infrared: \g / (b) Approximately 300 mg of 2—phenyl-1,3-dioxolane were dis- solved in 1 ml of bromotrichloromethane (Dow) contained in an nmr tube. This sample was immersed in a constant temperature bath maintained at 18° 90 and irradiated with a sunlamp (General Electric, lamp to sample distance = 3") for 3 hours. After this time the sample was then scanned on an nmr spectrometer and found to be quantitatively converted to 2-brom0ethy1 ben- zoate. The spectra for this sample were found to be identical in every respect to the infrared and nmr spectra obtained for the authentic product obtained according to method (a). 6. Preparation of 2-Bromoethylflluorobenzoate: p-Fluorobenzoyl chloride (Aldrich, 31.8 g, 0.2 mole) and 2-branoethanol (25 g, 0.2 mole) were combined in 40 m1 of carbon tetrachloride to give a light yellow, homogenous reaction mixture. This was refluxed for 12 hours, after which the solvent was removed on a rotating evaporator (Biichi). Distillation of the liquid residue through a 1/2 x 12" Vigreux column gave a major frac- tion boiling 95-100°/2 mm. Upon standing this distillate crystallized to an off-white solid, mp 38-43°. Redistillation of this crude product Save 20. 22 g (41%) of purified material which melted at 41-42.5° . Anal: Calculated for C9H8BrF02: C, 1:38; H, 3.26. Found: C, 44.0; H, 3.38. Infrared: i=0, 1720 cm-1 (Fluorolube) le': Spectrum 24 7- Preparation of 2-Bromoethpr;Chlorobenzoate: Into a 250 ml round-bottomed flask equipped with a reflux condenser was charged 52.5 s (0'3 mole) of p-chlorobenzoyl chloride (Hooker Chemical) and 37.5 g (0'3 mOle) of 2—bromoethanol. While being stirred with a magnetic stirrer, the r e8-<:tion mixture was heated at loo-125° for 12 hours. During this time the slightly viscous reaction mixture turned dark brown. After the mlxture was filtered to remove some suspended solid material, the filtrate was distilled through a 1/2 x 21" Vigreux column to give a major colorless 91 fraction coming over at 120-12l°/0.9 mm or 99-lOO°/O.2 mm which was prone to crystallize in the condenser and receiver (33.3 g, 68%, mp 28-32°). This product was recrystallized from diethyl ether, using a Dry Ice®- methylene chloride cooling bath, to give a fine white powder, mp 32.5-33.5° . gal: Calculated for C9H8BrC102: C, 41.0; H, 3.06. Found: C, 41.3; H, 3.16. Infrared: , 1732 cm'1 (neat) 52.0 LIE-E: Spectrum 25 8. Preparation of 2- Chloroethyl p-Chlorobenzoate: a—Chlorobenzoyl chloride (HOOker Chemical Company, 52.5 g, 0.3 mole) and 2-chloroethanol (24.15 g, 0.3 mole) were combined in a 150 ml round-bottomed flask equipped with a reflux condenser. The reaction mixture was stirred with a magnetic stirrer as the temperature was gradually increased. At a tem- perature of 85-100° , the evolution of hydrogen chloride was detected. The temperature of the reaction mixture was maintained at loo-110°, with stirring, for 12 hours. Distillation of this material through a 1/2 x 21" Vigreux column gave a water-white, major fraction boiling at 88-89°/1.5 mm. The product weighed 55.5 g (85%). Anal: Calculated for C H8C12 2: C, 49.3; H, 3.68. Found: C, 9 49.2; H, 3.86. Infrared: %=O’ 1730 cm-1 (neat) Nmr: Spectrum 26 9. Preparation of 2-Bromoethyl ELBromobenzoate: Ip-Bromobenzoyl chloride (Eastman, 43.9 g, 0.20 mole) and 2-bromoethanol (25 g, 0.20 mole) were combined in 50 m1 of carbon tetrachloride and refluxed for 13 hours. Solvent was removed at reduced pressure on a rotating evaporator (Bfichi) 92 leaving an amber liquid residue which weighed 58.9 g (96%). This crude material was distilled through a 1/2 x 12" Vigreux column giving a major cut (colorless) which boiled at 113-120°/2 mm and weighed 40.75 g (66%). This fraction was redistilled to give an analytically pure sample which boiled at 118-119°/2 mm. 2221? Calculated for C9H8Br202: C, 35.1; H, 2.62. Found: C, 35.2; H, 2.52. Infrared: 1738 cm.-1 (neat) ‘\7 /C=O’ Nmr: Spectrum 27 10. Preparation of 2-BromoethylApfFluorObenzoate: erluorObenzoyl chloride (Columbia, 11.9 g, 0.075 mole) and 2-bromoethanol (9.4 g, 0.075 mole) were combined in 15 ml of carbon tetrachloride contained in a 50 m1 round-bottomed flask equipped with a reflux condenser. The reaction mix! ture was refluxed for 18 hours during which time it darkened. Solvent was removed from the mixture on a rotating evaporator (Buchi) to yield a dark brown oily residue. This material was distilled through a 1/2 x 12" Vigreux column to give a major cut boiling at 89.5-90°/2 mm (14.3 g, 77%). 8221: Calculated for C9H8Br02F: C, 43.8; H, 3.26. Found: C, 43.6; H, 3.17. Infrared:-\ 1736 cm-:L (neat) 7 ,C=0’ Nmr: Spectrum 28 11. Preparation of 2-Bromoethy1 3L4-DichlorObenzoate: 3,4-Dichloro- benzoyl chloride (Heyden, 62.9 g, 0.3 mole) and 2-bromoethanol (37.5 g, 0.3 mole) were combined in 50 m1 of carbon tetrachloride and refluxed for 6 hours. The reaction mixture was cooled to ice temperature and fil- tered free of some suspended solid material. Solvent was removed on a 93 rotating evaporator (Bfichi) to give a liquid residue (80.5 g) which was distilled through a 1/2 x 21” Vigreux column. The distillate came over as a mixture of a colorless liquid and a white solid, bp l20-l25°/1 mm. The distillate solidified to a white mass in the receiver, mp 46-49°. Re- crystallization from a mixture of n-hexane and diethyl ether using a Dry IceD-methylene chloride cooling bath, gave a white crystalline product, mp 48-49.5°. apap: Calculated for C9H7BrC1202: C, 36.3; H, 2.37. Found: C, 36.1; H, 2.30. Infrared: :6=0’ 1737 cm.-1 (Fluorolube) Nmr: Spectrum 29 12. Preparation of 2-BromoethylpiTrifluoromethylbenzoate: prTri- fluoromethylbenzoyl chloride (Columbia, 10.4 g, 0.05 mole) and 24bromo- ethanol (6.25 g, 0.05 mole) was combined with 15 ml of carbon tetrachlor- ide in a 50 ml round-bottomed flask equipped with reflux condenser. The reaction mixture was refluxed for 24 hours after which the solvent was removed on a rotating evaporator (Bachi). A dark'brown residual oil was Obtained and fractionated through a 1/2 x 12" Vigreux column. A major cut, boiling at 82-88°/2 mm.was collected and redistilled to give 9.65 g (65%) of a colorless product boiling at 87-88°/2 mm. (aaap: Calculated for ClOHBBrF302: C, 40.3; H, 3.04; Br, 26.8. Found: C, 40.6; H, 2.95; Br, 26.6. Infrared: )6=0’ 1736 cm"1 (neat) Nmr: Spectrum 30 94 13. Preparation of 2-Bromoethyl p:_Trifluoromethylbenzoate: p-Tri- fluoromethylbenzoyl chloride (Columbia, 10.4 g, 0.05 mole) and 2-bromo- ethanol (6.25 g, 0.05 mole) were combined with 30 ml of carbon tetrachlor- ide in a 50 ml round—bottomed flask equipped with a reflux condenser. The reactants were refluxed for 9 hours and the solvent was then removed on a rotating evaporator (Blichi) to leave an amber liquid residue. Dis- tillation of this material through a 1/2 x 12" Vigreux column gave a major fraction boiling at 86-87°/1.5 mm (10.7 g, 72%). %: Calculated for ClOHBBrF302: C, 40.4; H, 2.71. Found: C, 40.5; H, 2.97. Infrared: :6=0, 1740 cm"1 (neat) Nmr: Spectrum 31 14. Preparation of 2-Bromoethyla-Nitrobenzoate: p—Nitrobenzoyl chloride (Eastman, 55.7 g, 0. 3 mole) and 2-bromoethanol (37.5 g, 0.3 mole) were combined in 50 m1 of carbon tetrachloride to give a deep red reaction mixture which was refluxed for 10 hours. When the mixture was cooled to room temperature, a red-yellow solid fell out of solution and was filtered. Solvent was removed from the filtrate on a rotating evap- orator (Blichi) to give a viscous yellow-red liquid residue. The liquid was fractionated through a 1/2 x 21" Vigreux column. A cut boiling at l30—l35°/2.5 mm was identified as the desired ester by nmr spectroscopy. Redistillation of this crude product gave a major fraction at 133-135°/ 2-5 mm (BO-75 s, 25%)- 1_\_n_a_l: Calculated for C9H8BrN04: C, 39.4; H, 2.94. Found: C, 39.7; H, 2.78. Infrared: :gzo, 1735 cm.1 (neat) Nmr: Spectrum 32 95 15. Preparation of 2—Bromoethyl p—Nitrobenzoate: p—Nitrobenzoyl chloride (Eastman, 55.68 g, 0.30 mole) and 2-bromoethanol (37. 5 g, 0.30 mole) were combined in 50 m1 ofcarbon tetrachloride and refluxed for 9 hours. The gray reaction mixture was cooled to room temperature and fil- tered free of some suspended solid material which was identified as 4-nitro- benzoic acid. Solvent was removed from the filtrate on a rotating evapor- ator (Blichi) to yield a gray solid product melting at 50-53° (46.45 g, 50%). Recrystallization from diethyl ether, using a Dry Ice®-methylene chloride cooling bath, gave a white powdery product melting at 49.5- 51° . A__nal: Calculated for C9HBBrN04: C, 39.4; H, 2.94. Found: C, 39.6; H, 3.07. Infrared: $4), 1730 cm“1 (Fluorolube) m: Spectrum 33 16. Preparation of 2-Bromoethyl Isophthalate: Isophthaloyl chloride (Eastman, 60.9 g, 0.3 mole) and 2-bromoethanol (75.0 g, 0.6 mole) were com- bined with 50 m1 of carbon tetrachloride and refluxed for 19 hours. After the mixture was cooled to room temperature, some white solid crystallized out of solution and was filtered. It was identified as isophthalic acid. Solvent was removed from the filtrate on a rotating evaporator (B'uchi) to yield a tan oily residue. This residue partially crystallized after standing in a stoppered flask for several days and was filtered, and washed with 20 ml of cold ether. The crude tan solid product weighed 45.13 g (39%) and melted at 50-56° . Recrystallization from a mixture of n-hexane and diethyl ether gave a nice white powder which melted at 51- 54° . The melting point was improved to 54-56° by placing the material in a vacuum desiccator over paraffin shavings and evacuating to 2 mm. 96 Anal: Calculated for 012812Br20h: C, 37.9; H, 3.18. Found: C, 38.1; H, 3.19. Infrared: \(73-0’ 1730 cnl':L (Fluorolube) --———— / _. m: Spectrum 34 17. Preparation of 2-Bromoethyl Terephthalate: Terephthaloyl chloride (Eastman, 60.9 g, 0.3 mole) and 2-bromoethanol (75 g, 0.6 mole) were combined in 50 m1 of carbon tetrachloride and refluxed for 5 hours. While the reaction mixture was being filtered to remove a small amount of suspended solid material, it solidified to a gray slushy mass. After it was cooled in an ice bath, the crude gray diester was filtered and found to weigh 80.9 g (71%), mp 90-94°. Several recrystallizations from di- ethyl ether gave a fluffy white product melting at 95-96° . _A_n_a_l: Calculated for Cl2H12Br2O4: C, 37.9; H, 3.18; Br, 42.1. Found: C, 38.0; H, 3.11; Br, 41.7. Infrared: i=0, 1733 cm’1 (Fluorolube) Nmr: Spectrum 35 18. Preparation of Tris (2-bromoethyl) Trimesate: Trimesoyl chlor- ide (Frinton labs, 26.55 g, 0.1 mole) and 2-bromoethanol (37. 5 g, 0.3 moles) were combined in 50 m1 of carbon tetrachloride. The homogeneous reaction mixture was refluxed for 12 hours, after which the solvent was removed at reduced pressure on a rotating evaporator (Bilehi). A viscous, tan oil remained which could not be induced to crystallize. An attempt to distill this material at a pot temperature of 175° at 2 mm was unsuc- cessful. A small amount of volatile material was collected, but the main portion would not distill under these conditions. The above treatment converted the material into a dark brown syrup which was soluble in diethyl ether and carbon tetrachloride but not in n-hexane. Adding 97 n-hexane to an ether solution of the syrup gave a white crystalline material which weighed 35.7 g (67%) and melted at 86—9l°. Recrystalliza- tion from diethyl ether (Dry Ice®-methylene chloride cooling bath) gave glittering white crystals melting at 94.5-96.5°. The melting point was increased to 95.5-96.5° by placing the material in a vacuum desiccator over paraffin shavings and evacuating to 5 um. A_nal_: Calculated for C15H15Br306: C, 33.9; H, 2.85. Found: C, 33.73 H: 2.61. Infrared: 7 0’ 1745 cm-1 (Fluorolube) >C= &: Spectrum 36 D. Precursors to 2,2'-A1ky1ene-l, 3-dioxolenium Dications 1. Prgparation of Bia(2-bromoeml) Oxalate: A solution of oxalyl Chloride (Eastman, 25 g, 0.20 mole) in 50 m1 of carbon tetrachloride was added dropwise with stirring over a period of 30 minutes to a solution of 2-bromoethanol (49.3 g, 0.39 mole) in 50 m1 of carbon tetrachloride. The reaction was not exothermic. While being stirred the reaction mixture was refluxed for 9 hours, after which the solvent was removed on a rotat- ing evaporator (B33chi). A viscous oily residue remained. Allowing this oil to stand in the hood draft caused some crystals to form on the walls of the evaporating dish. When these crystals were added to the oil and scratched, crystallization to a mass of gray platelets occurred. This crude material was recrystallized once from n-hexane to give 58.1 g (96%) of a glittering white crystalline product which melted at 54.5-56°. Cort and Pearson55 reported a melting point of 55-55.5°. 5%: Calculated for C6H8Br204: C, 23.7; H, 2.65. Found: C, 23.9; H, 2.65. Infrared: 1770, 1745 cm"1 (Fluorolube) \7 ) ,C=0 Nmr: Spectrum 12 98 2. Praparation of Bis- (2:bromoethyl) Malonate: 2-Bromoethanol (50 g, 0.4 mole) in 75 ml of carbon tetrachloride was added drOpwise to a stirred solution of malonyl chloride (28.2 g, 0.2 mole) in 75 ml of car- bon tetrachloride over a period of 30 minutes. A moderately exothermic reaction was observed. The reaction mixture was then refluxed, with stirring, for 6 hours, after which the solvent was removed at reduced pressure on a rotating evaporator (Biichi). The dark brown, viscous residue was distilled through a 1/2 x 12" Vigreux column giving a major colorless fraction boiling at 125-126° /0.5 m and weighing 47 g (74%). A__n_a_l: Calculated for (3731031204: C, 26.4; H, 3.17. Found: C, 26.4; H, 3.13. Infrared: i=0, 1761 cm-1 (neat) Nmr: Spectrum 13 3. Preparation of Bis (2-bromoethyl) Succinate: Succinyl chloride (Eastman, 26.5 g, 0.17 mole) and 2-bromoethanol (42.5 g, 0.34 mole) were combined in 75 ml of carbon tetrachloride and refluxed for 14 hours. Solvent was removed at reduced pressure on a rotating evaporator (Biichi) leaving a tan liquid residue which weighed 55.2 g (97%). The crude ester was taken up in 50 ml of carbon tetrachloride and washed with 125 ml of a saturated solution of sodium bicarbonate. The aqueous phase was ex- tracted with carbon tetrachloride (2 x 25 ml). The combined extracts were washed with water (2 x 100 ml) and dried over anhydrous calcium sul- fate. Solvent was removed at reduced pressure on a rotating evaporator (Buchi), leaving a nearly colorless liquid residue which weighed 37.8 g (67%) . This material was distilled through a 1/2 x 12" Vigreux column, yielding a colorless, major fraction which boiled at l45-l47°/2 mm and 99 weighed 27.3 g (48%). gasp: Calculated for C8H12Br20h: C, 28.9; H, 3.64. Found: C, 29.2; H, 3.92. Infrared: i=0, 1740 cm.1 (neat) m: Spectrum 14 4. Preparation of Bis- (2—bromoeglyl) Glutarate: Glutaryl chloride (Aldrich, 33.8 g, 0.2 mole) and 2-bromoethanol (50.0 g, 0.4 Mole) were combined in 75 ml of carbon tetrachloride and refluxed for 12 hours. Solvent was removed on a rotating evaporator (Bachi). The residue was distilled through a 1/2 x 12" Vigreux column and collected as a fraction boiling at 113-123° /2. 5 mm. Weight of the crude product was 51.2 g (74%). Redistillation of the crude product gave an analytical sample boiling at 115-118° /2. 5 mm. A_p_a_1_.: Calculated for 0931431204: C, 31.2; H, 4.08. Found: C, 31-33 H, 3-85- Infrared: >734), 1749 cm.1 (neat) Nmr: Spectrum 15 5. Preparation of His-(e-bromoetpyl) Adipate: 2-Bromoethanol (50 g, 0.4 mole) in 50 ml of Carbon tetrachloride was added dropwise to a stirred solution of adipyl chloride (36.6 g, 0.2 mole) in 50 m1 of the same sol- vent over a period of 30 minutes. A slight rise in temperature was noted (to 35°). The reaction mixture was stirred while at reflux for 17 hours, after which the solvent was removed at reduced pressure on a rotating evaporator (B'uchi). A light tan liquid residue was obtained which weighed 70 g (97%). This residue was distilled through a 1/2 x 12" Vigreux column, giving a colorless major fraction which boiled at 162-163° /0.9 mm and 100 weighed 52.80 g (73%). A_aa_1_: Calculated for ClOHl6Br20h: C, 33.4; H, 4.48. Found: C, 33.3; H, 4.48. Infrared: :g=0, 1749 cm.1 (neat) _1\Iz_nr_: Spectrum 16 6. Preparation of Bis(24bromoethyl) Pimelate: thelyl chloride (Frinton Labs, 19.7 g, 0.10 mole) and 2-brcmoethano1 (12.5 g, 0.10 mole) ‘were combined in 75 m1 of carbon tetrachloride and refluxed for 8 hours. Solvent was removed on a rotating evaporator, to give a dark'brown oil weighing 35.7 g (95%). This crude product was distilled through a 1/2 x 12" Vigreux column, yielding a light tan liquid as a major fraction which weighed 25.0 g (67%) and boiled at 150-159°/2 mm. This material was redistilled to give a pure product which boiled at 156-159°/2 mm. 522;: Calculated for C11H18Br204° C, 35.3; H, 4.85. Found: C, 35.5; H, 4.86. Infrared: :6=0’ 1750 cm.-1 (neat) flap; Spectrum 17 7. Preparation of Bis(2-bromoethyl)psuberate: Suberyl chloride (Frinton Labs, 21.1 g, 0.10 mole) and 2-bromoethano1 (24.9 g, 0.20 mole) 'were combined in 75 m1 of carbon tetrachloride. A vigorous reaction accompanied by a copious evolution of HCl was Observed shortly after addition. The reaction.mixture was refluxed for 25 hours, after which the solvent was removed at reduced pressure on a rotating evaporator (Bfichi). A heavy, tan liquid residue remained and crystallized into a tan solid upon standing overnight at room.temperature. The crude product 'weighed 26.4 g (94%) and melted at 31-35°. Recrystallization from 101 n-hexane gave glittering white plates which melted at 37-38° . Anal: Calculated for 012H2OBr2O4: C, 37.1; H, 5.19. Found: C, 37°33 H: 5°21- Infrared: \6-0’ 1750 cm.1 (Fluorolube) —— I _ Ling: Spectrum 18 E. 2-Alky1-1, idioxolenium Cations: 1. Preparation of 2- (N,N-Dietpy_1amino)-l, 3-dioxolenium Tetrafluoro- b_0papa: Under anhydrous conditions, powdered silver tetrafluoroborate (1.93 g, 0.01 mole) was added in one portion to a stirred solution of 2- bromoethyl N,N-diethy1carbamate (1.96 g, 0.01 mole) in 20 ml of dry methylene chloride. An exothermic reaction sufficient to cause the sol- vent to reflux slightly was observed and was accompanied by the formation of a yellow precipitate. The reaction mixture was stirred at 25-30° for 1 hour after which the insoluble material was filtered, washed with methylene chloride (2 x 10 ml) and found to weigh 1.64 g (88%, assuming it is only silver bromide). Reduction of the filtrate to one-half its original volume and addition of diethyl ether caused a colorless oil (lower phase) to separate. This layer could not be induced to crystal- lize until it was cooled in a Dry Ice®-methy1ene chloride cooling bath and scratched. After some effort, a white crystalline material formed and was filtered. It weighed 1.6 g (69%), mp 46-51°. Recrystallization of this crude material from methylene chloride (Dry Ice® cooling) yielded fine white crystals melting at 53-54.5°. A_na_1: Calculated for C7314NO2.BF4: C, 36.4; H, 6.11. Found: C, 36.4; H, 6.29. m: Spectrum 37 102 2. Preparation of 2— (gyc1opr02111- 1, 3- dioxolenium Tetrafluoroborate: Under anhydrous conditions, powdered silver tetrafluordborate (1.93 g, 0.01 mole) was added in one portion to a stirred solution of 2—bromoethyl cyclopropanecarboxylate (1.93 g, 0.01 mole) in 20 ml.of dry methylene chloride. An.immediate precipitation of a yellow cream-colored solid accompanied the slightly exothermic reaction. {After the reaction.mixture was stirred for 2 hours at 25-30° with a magnetic stirrer, the suspended solid was filtered and washed with.methylene chloride (2 x,lO ml). The yellow-gray filter cake weighed 3.30 g (85%, assuming it consists of cation and silver bromide). Reducing the filtrate to half the original volume and adding 10 m1 of diethyl ether gave 0.05 g (2.5%) of the cation salt as a white crystalline product which melted at 123-125.5°. The above filter cake was slurried in 10 ml of liquid sulfur dioxide, filtered and then washed 'with liquid sulfur dioxide (2 x 5 m1). A yellowagray filter cake of sil- ver bromide remained which weighed 1.55 g (83%). Evaporation of the fil- trate yielded 1.25 g (63%) of a white crystalline product. This material was recrystallized from a mixture of acetonitrile and methylene chloride (Dry Ice‘E—methylene chloride cooling bath) to give fine white crystals melting at l2h.5-l26°. Agal: Calculated for C6H9O2'BFA: C, 36.0; H, b.5h. Found: C, 35.8,- H, n.3u. Nmr: Spectrum 38 3. Preparation of 2-(2—Methylprgpepyl)-l,3-dioxolenium.Tetrafluoro- ‘borate: Anhydrous powdered silver tetrafluoroborate (1.93 g, 0.01 mole) 'was added in one portion to a stirred solution of 2-bromoethyl 3,3-dimethyl- acrylate (2.07 g, 0.01 mole) in 20 ml of dry methylene chloride while under anhydrous conditions. The reaction was mildly exothermic and accompanied 103 by the formation of a creameyellow precipitate. The reaction mixture was allowed to stir for 2 hours at 25-30°. After this time the precipitate was filtered and washed with methylene chloride (2 x 10 ml). The filtrate was reduced to dryness leaving a slushy gray solid. This residue was dis- solved in 5-10 ml of methylene chloride and filtered. Anhydrous diethyl ether was added; this caused the product to precipitate. The crude product was filtered and found to weigh l.h g (65%), mp 57-6l°. Recrystallization from a mixture of diethyl ether and methylene chloride (Dry Ice‘E cooling) gave a fine white crystalline product melting at 6l-62°. Anal: Calculated for C7H1102'BFh: C, 39.3; H, 5.18. Found: C, 39.0; H, 5.03. Egg: Spectrum 39 h. Preparation of 2-(p-Methoxygtyryl)-l,3rdioxolenium.Tetrafluoro- borate: 2-Bromoethyl h-methoxycinnamate (2.85 g, 0.01 mole) in 20 ml of methylene chloride was charged into a 50 ml erlenmeyer flask equipped with a magnetic stirrer. In a dry box, (1.93 g, 0.01 mole) of powdered silver tetrafluoroborate (Alfa Inorg) was added in one portion to the stirred ester solution. A precipitate formed immediately and the solution became a brilliant canary yellow. The reaction mixture was stirred for 1 hour at 25-30°, after which the yellow precipitate was filtered (3.h g, 71% assume ing it consists of the cation and silver bromide). Trituration of the precipitate with 10 ml of liquid sulfUr dioxide followed by filtration and “washing the filter cake with liquid sulfur dioxide (2 x 10 ml) left a creamecolored filter cake of silver bromide (1.65 g) and a brilliant yellow filtrate. Evaporation of the sulfur dioxide yielded a yellow solid ‘Which weighed 2.08 g (71%). This material turned dark at.e130-1h0° and decomposed at 185-195°. Recrystallization from.a mixture of acetonitrile 10h and methylene chloride using a Dry Ice®-methylene chloride bath, gave a brilliant yellow powder which darkened at4v210° and decomposed to a black resin at 220-223°. This product gave a canary yellow solution in liquid sulfur dioxide, methylene chloride or acetonitrile but was rapidly decol- orized in FSOBH or water. Anal: Calculated for C12Hl3O3OBFh: C, h9.5; H, h.30. Found: C, h9.5; H, h.h5. Nmr: Spectrum.hl 5. Preparation of trans-2-(Propenyl)-1,3-dioxolenium Tetrafluoro- bggatg: under anhydrous conditions, powdered silver tetrafluordborate (1.93 g, 0.01 mole) was added in one portion to a stirred solution of 2- bromoethyl crotonate (1.93 g, 0.01 mole) in 20 ml of anhydrous methylene chloride. A yellow precipitate formed shortly after the addition. The reaction.mixture was stirred for 1 hour at 25-30° after which time the insoluble material was filtered and washed with methylene chloride (2 x 10 ml). The yellow-gray filter cake weighed 3.38 g. This represents a yield of 87%, assuming it consists of cation salt and silver bromide. This filter cake was slurried in.lO ml of liquid sulfur dioxide, filtered and ‘washed twice with 5-ml portions of this solvent. The silver bromide re- maining on the filter weighed 1.38 g (9h$). Reducing the filtrate to dryness gave 1.58 g (79%) of a fine crystalline material.melting at 1&0- lh7°. Recrystallization from a mixture of acetonitrile and methylene chloride gave an analytical sample of glittering white crystals melting at 150-152°. Anal: Calculated for C6H 902°BFh: C, 36.0; H, h.5h. Found: C, 36.3; H, h.hl. Nmr: Spectrum #0 105 6. Preparation of 2-(Styryl)-1,3rdioxolenium.Tetrafluoroborate: Powdered silver tetrafluoroborate (1.93 g, 0.01.mole) was added in one portion to a stirred solution of 2—bromoethyl cinnamate (2.65 3,0.01 mole) in 20 ml of methylene chloride under anhydrous conditions. The reaction mixture was stirred for 1.75 hours, filtered, then washed with methylene chloride (2 x 10 ml). The yellow gray filter cake weighed 3.2 g (77%, assuming it consists of silver bromide and cation salt). This filter cake was slurried in 10 m1 of liquid sulfur dioxide, filtered and washed twice with 5 ml portions of the solvent. A filter cake of silver bromide re- mained which weighed 1.7 g (92%). The sulfur dioxide filtrate was re- duced to dryness, to yield 1.45 g (67%) of an offbwhite solid product which melted at l73-176°. Recrystallization of this crude material from a mix, ture of acetonitrile and methylene chloride (Dry IcéE cooling) gave glit- tering white crystals which melted at 178—179.5°. Anal: Calculated for CllH1102OBF4: C, 50.4; H, 4.23. Found: C, 50.1; H, 4.34. EEE‘ Spectrum 42 7. Preparation of 2-(ErButy1)-1,3-Dioxolenium.Tetrafluoroborate: Under anhydrous conditions powdered silver tetrafluoroborate (Alfa Inorgan- iC, 1.93 g, 0.01 mole), was added in one portion to a solution of 2-bromo- ethyl pivalate (2.09 g, 0.01 mole) and 15 m1 of anhydrous methylene chlor- ide contained in a 50 m1 erlenmeyer flask equipped with a magnetic stirrer. An immediate exothermic reaction was Observed accompanied by the formation of light yellow precipitate. The flask was stoppered and the reaction.mix- ture‘was stirred for 2 hours at 25-30°. After the silver bromide was fil- tered and washed with methylene chloride (2 x 10 ml), the filter cake 106 weighed 1.8 g (96% of theory for AgBr).. When.the filtrate was cooled, some product crystallized out, but addition of a small amount of diethyl ether caused more complete crystallization. The product was filtered as a white fluffy solid melting at 148-151° (1.55 g, 72%). An analytical sample was obtained by dissolving the product in a 5:1 methylene chloride:acetonitrile mixture and adding a small amount of diethyl ether. Cooling in a Dry Icém bath gave fine white needles, melting at l5l.5-152.5°. Anal: Calculated for C7H1302OBF4: C, 38.9; H, 6.07. Found: C, 38.9; H, 6.12. Ann: Spectrum 44 8. Praparation of 2-(Cyclobutyl)-l,3-dioxolenium.Tetrafluoroborate: Under anhydrous conditions, powdered silver tetrafluordborate (Alfa In- organic, 1.93 g, 0.01 mole) was added in one portion to a stirred solution of 24bromoethy1 cycldbutanecarboxylate (2.07 g, 0.01 mole) in 15 m1 of anhydrous methylene chloride. Although the immediate f0rmation of a light yellow precipitate was Observed, the reaction did not appear to be exo- thermic. The reaction mixture was stirred for 1 hour at 25-30°. The sus- pended silver bromide was filtered and washed twice with lO-ml portions of methylene chloride. The weight of filter cake'was 1.85 g (99%). The colorless filtrate was reduced to 1/2 its Original volume under reduced pressure and then anhydrous diethyl ether was added until a colorless oil separated from.the solution. The two-phase system was cooled in a Dry Icém- methylene chloride bath and scratched. After some effort, the oil finally crystallized to a white fluffy solid which was filtered, washed with 10 ml of ether and faund to weigh 1.20 g (56%), mp 28-3l°. Recrystallization from a mixture of methylene chloride and diethyl ether (Dry Ice(s) 4.44 (t); Jca 6.5 6.23 (d); Jd8 16.2 6.83 (d); Jef 8-9 7.43 (d); er 8.9 7.60 (d); J8d 16.2 o‘ a: It 0 a. o o' a x a: :c:d:e:(f+g) = 2:3:2:1:2:3 133 Spectrum.6. Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1 Cinnamate (0011‘) (d) (a) (e) H C-C-O~CH2-CHé—Br (d) (d) 1': (b) (a) (c) (d) (b) (0.) (X) «a an Aaalgnments (Q) 0.00 (TMB) 3-53 (t); Jab - 6.0 cps 4.46 (t), . 6.37 (d); Jce a 16.2 cps 7.42 On) 7.66 (d) :c:(d+e) - 2:2:l:6 dun-060x 134 Spectrum 7. Nuclear Magnetic Resonance Spectrum of 2-Bromoethyl Cyc10buty1carboxy1ate (CClh) 9 (a) (c) 7., ,Qwaa H (b) (d) (c) (X) (a) (b) Assignments (Q) x 0.00 (TMB) a 2.14 (m) b 34MB 0n) c 3.49 (t); ch = 6.0 cps d 4. 36 (t) a:b :c:d = 3:1:2:2 135 U Spectrum 8. Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1 Pivalate (0011‘) (a) 983.9 (C) (b) (a) wB-a—C-ocse-aia-sr (a) CH3 (0.) (c) (b) L4. 1 Assigments (0) 0.00 (ms) 1.20 (s) 3.50 (t); ch = 6.0 cps 4.35 (t) a:b:c = 9:2:2 00'9“ Spectrum 9 . Nuclear Magnetic Resonance Spectrum of 136 2-Bromoethy1 Acetate (CClh) (a) CH 9 (c) (b) I 3- C- 0- CH2- (152-Br (a) tn. 09 oc‘mx a:b:c Asslgnnents (Q) 0.00 (ms) 2.06 (s) 3.50 (t); ch = 6.0 cps 4.35 (t) = 3:2:2 (x) 137 Spectrum 10. Nuclear Magnetic. Resonance Spectrum of 2-Bromoethyl Methacrylate (CClh) (d) H\C_C/CH3 (a) (e) H/ \C-O-CH2-CH2-Br " o (c) (b) (a) (c) (b) (x) (0) (cl) Assignments (0) x 0.00 (1148) a 1.96 (8) b 3.54 (t); ch = 6.0 cps c 4.43 (t) d 5.61 (m) e 6.13 (s) °b:c:e:d = 3:2:2:l:1 138 Spectrum 11. Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1 Acrylate (001“) (c) ”Nae/H (a) (d) H/ \E-O-CHZ-CHQ-Br 0 ('0) (a) (b) (a) (d) r-‘H (c) (x) (‘5 Assigments (0) .0.00 (ms) 3.52 (t); Jab = 6.0 cps 4.44 (t) 5-68-5-99 (In) 6.12-6.65 (m) 39060:: :c:d = 2:2:1:2 139 Spectral: 12. Nuclear Magnetic Resonance Spectrum of Bis-2-bromoethyl Oxalate (001”) (80.60) 99 (b) (3) Br- 032- GHQ-O-C- C-O-CHQ- CHz-Br (x) (b) (a) Aaalggnents (6) 0.00 (ms) a 3.59 (t); Jab . 6.2 cps b 4.59 (t) a:b :- 1:1 140 Spectrum 13. Nuclear Magnetic Resonance Spectrum of Bis (2-bromoethyl) Malonate (001“) 0 0 n n Br- CH2- GHQ-0- C- CH2- C- 0- 052- 0112- Br (b) (e) (a) (e) (b) (a) «a 00 (b) Wants (51 0.00 (m) 3.45 (a) 3.54 (t); ch = 6.0 cps 4.45 (t) (3de (a+b):c = 3:2 141 Spectrum 14. Nuclear Magnetic Resonance Spectrum of Bis(2-bromoethyl) succinate (0011‘) 0 . 0 Br-CH2- 0112-030- 032- one-C-o-CHQ-Csz-Br (b) (e) (a) (a) (c) (b) (a) (c) (b) (X) Ailments 15) x 0.00 (EMS) a 2.65 (s) b 3.52 (t); ch = 6.0 cps c 4.40 (t) a:b:c - 1:1:1 144 t 1 . Ma ti R Spec rum 7 Bis(gf§rangg§hyl) Igggglggg i8}??? of o Br-CHZ- ~2CH -0-C- ”one CLHz- CH2- CH2- CH2- -0- ---0-CH2 0112- -Br (c) (d) (b) T (b) (d) (c) (a) (c) (d: (b) (a) (x) H-x Assignments (p) x 0.00 (ms) a 1.09-2.00 (m) b 2.06-2.54 (m) c 3.49 (t); Jed = 6.0 cps d 4.35 (t) a:b:c:d = 3:2:2z2 145 Spectrum 18. Nuclear Magnetic Resonance Spectrum of Bis(2-bromoethy1) Suberate 0 0 Br-CH2- cue-o-c- (mg- £112— 032- 0112- 035- Cfie-C-O-Cfie- CH2-Br (c) (d) Ch) 7 (b) (d) (c) (a) (x) (d) (c) (a) (b) r-H Analgments (6) 0.00 (THE) 1.45 (m) 2.32 (t); Jba '.‘ 6.5 cps 3.49 (t); ch = 6.0 cps 4.35 (t) a:b:c:d = 2:1:1:1 9:0de Spectrum 19. (a) (d) CDC-06'9” 146 Nuclear Magnetic Resonance Spectrum of 2—Bmmoethyl-p-methomenzoate (0011‘) (a) (e) 0 CH3..— C-O-CHQ- CHZ-Br (e) (a) (b) (d) (e) (x) (c) , (a) Amen” (91 0.00 (ms) 3.56 (t); Jae = 6.0 cps 3.79 (a) 4.52 (t) 6.84 (d); Jde a 8.7 cps 7.93 (e) a:b:c:d:e . 2:3:2:2:2 147 Spectral: 20. Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1-3, 4, 5- trimethoquenzoate (001,.) ((3) m3. (8) .9 (b) CH 0 -0-CH -CH -Br 3 < > 2 2 ' (d) (a) (c) 01130 (e) (b) (X) (a) (d) I (a) Analggents (Q) 0.00 (M) 3.60 (t); Jad = 6.0 cps 3.81 (s) 3.87 (a) I“55 (t) 7-19 (3) a:b:c:d:e s 2:3:6:2:2 09.0de 148 Spectrum 21. Nuclear Magnetic Resonance Spectrum of 2-Bromoethyl-p—methylbenzoate (0°14) (a) (e) o (s) 033. -.C-O-Cfiz-CH2-Br (c) (b) (d) (e) (a) (C) (b) (e) a!) (x) Ass cats 6 0.00 (TMB) 2-39 (0) 3.57 (t); ch - 6.0 cps 4-53 (t) 7.17 (d); Jde - 8.0 cps 7.89 (d) (DD-00'9" 149 Spectrm 22. Nuclear Magnetic Resonance Spectrum of 2-Brom0ethy1-n-methylbenzoate (0011‘) (a) (e) o f " -O-CH2- GHQ-Br (e) (b) CH3 (e) (a) (a) (x) (c) (t) (d) m (0) Assignnents (6) 0.00 (ms) 2.37 (a) 3-57(t)3 ch 3 6-0 QPB 4-54 (t) 7.20-7.44 (m) 7-82 (a) a:b:c:d:e .= 3:2:2:2:2 o n. o 0' m x 150 .Spectrumr 23. .Nu clear 2-3 Magnetic Resonan romoe Benzoate (03: Spectrum Of h d) ( 0 c) { " 0-CH2-CH , 1"” (a) (b (a m m (b) (a) (d) Asslgnents (6 ) : 0.00 (ms) 358 (t); J 2 ”.57 (t) ab = 6.0 cps d 7.40 (m) 8.02 (m) a:b:C°d . = 2:2'3 . :2 151 Spectrum 24. Nuclear Magnetic Resonance Spectrum of 2-Bromoethyl-p-f1u0robenzoate (0011‘) “1’ ‘°’ 0 (b) (a) n F —C-O-CH2-CH2-Br (d) (e) (x) (b) M (d) ‘0 Asslments (6) 0.00 (ms) 3.61 (t); Jab = 6.0 cps 4.60 (t) 7-12 (m) 8.08 (m) a:b:c:d == 1:1:1:1 9:060“ 152 Spectrum 25. Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1-p- chlorobenzoate (0011)) (c) (a) O Cl.—¢-O—CH2- One-Br (b) (a) (e) (d) (x) (c) (a) (b) (a) Assi ents 6 x 0.00 ('36) a 3.60 (t); Jab - 6.0 cps b “-57 (t) c 7.36 (d); Jed . 8.5 cps d 7.94 (a) a:b:c:d = 1:1:1:1 153 Spectrmn 26. Nuclear Mametic Resonance Spectrum of 2— Chloroethyl-n- chlorobenzoate (0011‘) (e) (d) o (c) r-ocna-an-Cl (c) (b) (a) (d) r‘o (x) Asslgnents (6) x 0.00 (1148) a 3.78 (t); Jab = 5-5 cP3 b 4.54 (t) c 7.40 (m) d 7.88 (m) a:b:c:d = 1:1:1:1 154 Spectrum 27. Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1-m-bromobenzoate (0011‘) (c) o fg-O-Cfle- GHQ-Br (e) (8) Dr (C) (b) (a) (c) _..¥ A __‘ w W vii fi v Assignments (6) x 0.00_ (SIMS) a 3.61 (t); Jab = 6.0 cps b 4.59 (t) c 7.12-8.19 (m) a:b:c = 1:1:2 (x) 155 Spectrum 28. Nuclear Magnetic Resonance Spectrum of 2-Bromoethyl-m- fluorobenzoate (CClh) (c OC-O—CH CHM-CH -Br (‘0) (a) F (Ge) 1 (b) m (c) (x) Assignments (6) x 0. 00 (ms) a 3- 52 (t);J = 6. 0 cps b 4. 61 (t) C 7~03-7-95 (In) a:b:c = 1:1:2 Spectrum 29. (c) :3de -/ 15:“) Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1- 3, 4-Dichlorobenzoate (0011)) (c) A) 0 a . awe-.. O Cl (c) (t) (a) (x) Asslgnments Q) 0.00 (ms) 3.60 (t); Jab = 6.0 cps 4.61 (t) 7.40-8.17 (m) a:b:c = 2:2:3 15.7 Spectrum 30. Nuclear Magnetic Resonance Spectrum of 2-Branoethyl-n- trifluoromethylbenzoate (001,.) (b) (A) (a) (c) (X) r-‘fi Msignlients (6) 0.00 (M) 3.63 (t); Jab a 6.0 cps 4.64 (t) 7.84-8.03 (m) 8.11-8.36 (m) 900'“): a:b:c:d a 1:1:1:1 158 Spectrum 31. Nuclear Magnetic Spectrum of 2-Br0moethyl- ' p-trifluoromethylbenzoate (001“) (e) d) 0 F30 —-C-0-CHZ- Gila-Br (b) (a) (c) (d) (X) (b) (a) (d) (c) Assi nts 6 0.00 (m) 3.64 (t); Jab a 6.0 cps 4.65 (t) 7.66 (d); J = 8.6 cps 8.14 (d) a:b:c:d - 1:1:1:1 DOUG“ Spectrum 32. (d) (d) «nrfia «n 'b c d 159 Nuclear Magnetic Resonance Spectrum of 2-Bromoethy1-n-nitrobenzoate (CClh) (ms-uh internal standard) (c) (a) o .. O-CHe-Cfia-Br (b) (a) 2 (e) (b) (a) Asslggents (6) 3.72 (t); Jab = 6.0 cps 4-70 (t) 7.68 (t); J .= 7.5 cps 8-36 (m) 8.70 (m) a:b:c:d:e a 2:2:1:2:l 160 Spectrum 33. Nuclear Magustic Resonance Spectrum of 2-Bromoethy1-p-nitrobenzoate (0011‘) (c) c) 0 ... ...... (c) (c) (b) m (c) (x) (b) (a) Assi nts 6 x 0.00 (nus) a 3.68 (t); Jab - 6.0 cps b 4.67 (t) c 8.23 (s) a:b:c - 1:1:2 161 Spectrum 34. Nuclear Mapetic Resonance Spectrum of 2-Branoethyl Isophthalate (cell) 0 (a) g-O-CHQ-CHZ-Br b) a (d‘k) ( ( ) (d) s-o-CH -cn -Br 2 2 0 (b) (a) (a) (a) (x) H) ' (o) As_s_:_Lgnnents (6) x 0.00 (M) a 3.64 (t); Jab s 6.0 cps b 4.63 (t) c 7.52 (m) d 8.24 (m) e 8.66 (m) a:b:c:d:e = 4:4:2:1:1 162 Spectrum 35. Nuclear Magnetic Resonance Spectrum of 2-sromoethy1 Terephthalate (0°14) (c) (e) sr-ma-CHa-o-o-CHZ-Csz-sr a b b ) ()() (0)“) ()(a (x) (c) (b) (a) Writs (61 0.00 (ms) 3.67 (t); Jab - 6.0 cps 4.68 (t) , 8.15 (s) a:b:c -- 1:1:1 OO’DN 163 Spectrum 36. Nuclear Magnetic Resonance Spectrum of 2-Br0moethy1 Trimesate (0011‘) 0 :2 ‘°’ "-Wa-wa-Br m-CHZ-CHa-O- 0 MM (3) (a) (b) (c) S—O—CHZ-Cfie-Br 0 (b) (a) (:0 (C) (b) (a) 119M141). x 0.00 (M) a 3.70 (t); Jab =- 6.0 cps b 4.73 (t) c 8.91 (s) a:b:c a 2:2:1 164 Spectrum 37. Nuclear Magnetic Resonance Spectrum of 2— (N, N- Diethylamino)-1, 3- dioxolenium TetrafluorOborate (FSO3H) (a) (a), (:0 Agar-gents (21 3.10 (TMAIBFh) 1.32 (t): Jab = 7.5 cps 3-59 (q) 4.98 (s) a:b:c = 3:2:2 oo‘mx 165 Spectrum 38. Nuclear Magnetic Resonance Spectrum of 2- (Cyclopropyl)-l, 3-dioxolenium Tetrafluoroborate (FSO3H) ‘°’ {Du-<3” mi. ‘ H (c) . (b) (c) Assignments (6) 3.10 (M’BFh) 1.75-2.07 (m) 2.10-2.53 (m) 5.17 (8) a:b:c = 4:1:4 oo‘mx 166 Spectrum 39. Nuclear Magnetic Resonance Spectrum of 2— (2-Methylpropenyl) -1, 3- dioxolenium Tetrafluoroborate (FSO3H) (b) (b) (a) (x) (c) Asslggugnts (6) x 3.10 (M'BFh) a 2-37 (d); Jae = 7-5 cps b 5.19 (a) c 6.27 (m) a:b:c = 6:4:1 167 Spectrum 40. Nuclear Magnetic Resonance Spectrum of trans-2- (Propemrl) -1, 3-dioxolenium Tetrafluoroborate (r3031!) (3) m3\c C/H (c) (d) H’ . o - (b) BF" (5) (x) (a) (d) (c) Writs (a) 3.10 (TMAIBFh) 2.27 (d); J . 7.0 cps; J = 1.5 cps 5-23 (8) 6.44 (m) 8.22 (m) a:b:c:d a 3:4:1:1 n-Ourox 168 ‘ Spectrum 41. Nuclear Magnetic Resonance Spectrum of 2- (p-Methoxystyrl)-1, 3- dioxolenium Tetrafluoroborate (~20$ in ISO 3H) (a') H (1.19-2: .. ... ....\<§jm W4 H (bu) (C') (b) (e) (b) (a) (x) ('3 c,c’ ! ’7 0' H—‘fi (45') Assignments (6) I II x 0.00 (EMA-BF“) x 0.00 (mu-ark) a 4.15 (s) a' 4.02 (s) b 5.29 (s) b' 5.22 (s) c,c' 6.74-8.67 (m) c,c' 6.74-8.67 (m) 169 Spectrum 42. nuclear Magnetic Resonance Spectrum of 2- (Styrl)-J., 3-dioanolen1un Tetrafluoroborate CEBO3H) ,s (d) M {<.> K661“) BF): fi <> (a) 0b) (an (e) (’0 ed) an Ass gents (6) 3.10 (EMA-mm) 5.25 (a) 6.86 (d); de ..-. 16.0 cps 7.35-7.93 an) 8.59 (d) a:b:c:d . 4:1:5zl 9069K 170 Spectrum 43. Nuclear Magnetic Resonance Spectrum of 2- (Cyclobutyl) -:L, 3- dioxolenium Tetrafluoroborate (FSO3H) o (c) ( ) <<::>K-<<+ BF - 6 <{ H ‘°::]@;) 4 (b) (c) W9). 3.10 ('IMA-BFL) 1.86-2.83 (m) 3-42-3-89 (q) 5.29 (s) a:b:c = 6:1:4 OO‘WN 17.1 Spectrum 44 . Nucl ear Magnetic Resonanc e Spectrum of 2- (t-But _ y1)-1 borate (Fs ’B-dioxu 03H) enium Tetr afluoro- (a) CH (3) CH -' 3 I (b) (a? 'CH 1+ BF“. 3 (b) (b) (ah a:b Assignments (5) 3.10 ('IMA'BF ) 1.50 (s) A 5-30 (a) = 9:4 3.72 Spectrum 45. Nuclear Magnetic Resonance Spectrum of 2- (Methyl) -1, 3- dioxolenium Tetrafluoro- borate (13033) (b) (a)CH<-+ BF- 3 \ 4 3(1») (b3 (0) (k) W x 3.10 (M'BFh) 2-75 (a) 5-30 (8) a:b = 3:4 173 Spectrum 46. Nuclear Magnetic Resonance Spectrum of (c) 00'.” 2- Isopropemrl- l, 3-‘dioxolenimn Tetra- fluoroborate (330311) 0113 (a) ( a {1K / ma" GUI.) “ (b) (a) (>0 W 3.10 (nun-ark) 2.15 (a) 5.3:. (a) . 6.85 (a) a:b:c -- 3:4:2 174 Spectrum 47. Nuclear Magnetic Resonance Spectrum of 2-(vm1)-1, 3- dioxolenium Tetrafluoro- borate (F8031!) (a) CK) (b3 Ass iggnents (61 x 3.10 (M'BF’I‘) 5.35 (s) b 6.28-7.57 (m) a:b = 2:1 175 Spectrum 48. Nuclear Magnetic Spectrum of Product Fran Bis (2-bromoethyl) Oxalate and Two Equivalents or Angl’6 (19033) H] 2 SbF6 and/or ago-011241124:- (b) (x) (c) (a) Assignnents (Q1 2: 3.10 (M'BFh) “-79-5-07 (m) b 5-70 (8) c 6.00 (s) .a:b:c = 5:l2:l 176 Nuclear Magnetic Resonance Spectrum of Spectrum 49 . 2, 2' -Methy1enebis-l, 3—dioxolenium Tetrafluoroborate (FSO3H) (EWIII’ ** -<<:::r0 - OCH l+ 23F (a) o 2 \ a) h (03 amen“ ('11 x 3.10 (TMAfBFh) 5.61 (s) * Exchanges with F803}! 177 Spectrum 50. Nuclear Magnetic Resonance Spectrum of 2, 2' -Ethylenebis- l, 3- dioxolenium Tetra- fluoroborate (F8033) (a) ‘ , (a) +1 CH -CH 0+ 2 BF " (a) ' (13 (b§0 G. A. Olah, C. A. Cupas and M. B. Comisarow, J. Am. Chem. Soc., J "‘ 22. 362 (1966)- G. A. Olah and C. U. Pittman, Jr., ihid, _88, 3310 (1966). G. A. Olah, R. C. Chambers and M. B. Comisarow, ibid, 89, 3586 (1957). D. M. Brouwer, Chem. Comm, 515 (1967). G. A. Olah and J. M. Bollinger, J. Am. Chem. Soc., Q, 3582, 3586 (1967). G(. .2.)01ah and J. M. Bollinger, J. Am. Chem. Soc., 89, 2993, 4744 19 7 . F0 Me Beringer and Be A. Gal-ton, J0 Org. was, 3% 2630 (1967). S. Kabuss, Angew. Chem. Internet. Edit. , 675, 6 (1966). K. Dimroth and P. Heinrich, ibid, 9, 677 1966). (.2. A.')01ah and C. U. Pittman, Jr., J. Am. Chem. Soc., 87, 3507 1965 . G. A. Olah, C. U. Pittman, Jr., E. Namonworth and M. B. Comisarow, ibidi 81. 5571 (1966). G. A. Olah, M. B. Comisarow, E. Namonworth and B. Ramsey, J. Am. Chem. Soc., 89, 5259 (1967). 121. Brookhart, F. A. L. Anet and S. Winstein, ibid, 88, 5657, 5659 1966). L. Eherson and s. Winstein, ibid, 87, 3506 (1965). G. A. Olah, J. Am. Chem. Soc., 87, 1103 (1965). G. A. Olah and T. E. Kiovsky, ibid, 89, 5692 (1967). R. w. Taft and L. D. McKeever, ibid, 87, 2489 (1965). B. 0. Ramsey and R. w. Taft, ibid, 88, 3058 (1966). 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35- 36. 37- 38. 39- 4o. 41. 42. 43. 44. 203 H. Meerwein, V. Hederich and K. Wlmderlich, Arch. Pharm., 291, 541 (1958). ‘S. Winstein, Hess 8c Buckles, J. Am. Chem. Soc., 6_4, 2780, 2787, 2796 (1942). s. Winstein and R. E. Buckles, J. Am. Chem. Soc., 82, 613 (1943). R. s. Tipson, J. Biol. Chem., 139, 55 (1939). H. s. Isbell, Ann. Rev. Biochem., 2, 65 (1940). W. W. Pigman and M. L. Wolfram, "Advances in Carbohydrate Chemistry," Academic Press Inc. , Publishers, New York, N.Y. Vol I, p 77 (1945). Ibid, V01 Ix, p 2 (1954). C. Prevost, Compt. rend., 126, 1129 (1933). C. Prevost and J. Wiemann, Compt, rend. , 204, 700 (1937). R. B. Woodward and F. V. Brutcher, Jr., J. Am. Chem. Soc., _8_9, 209 1958). G. E. McCasland and E. C. Horswill, J. Am. Chem. Soc., 16, 1654- (1954). C. A. Bunton and M. D. Carr, J. Chm. Soc., 770 (1963). C. B. Anderson, E. C. Friedrich and S. Winstein, Tetrahedron Letters, 255 2037 (1953). H. Meerwein, K. Bodenbenner, P. Borner, F. Kunert and K. Wunderlich, Ann, 632, 38 (1960). Cope and‘Wood, J. Am. Chem. Soc., 12, 3885 (1957). s. Winstein and R. Boschan, J. Am. Chem. Soc., 18, 4921 (1956). C. Pedersen, Tetrahedron letters, 8, 511 (1967). M. J. Coppen, M. P. Hartshorn and D. N. Kirk, J. Chem. Soc. (c), 575 (1966). C. F. Wilcox, Jr. and D. I. Neally, J. Org. Chem” g§, 3446 (1963). J. G. Buchanan, J. Chem. Soc., 2511 (1958). 3. Julia and J. P. Lavaux, Bull. Soc. Chim. Fr., 1238 (1963). J. M. Coxon, M. P. Hartshorn and D. N. Kirk, Tetrahedron Letters, 22, 2547 (1964). J. G. Buchanan and R. Fletcher, J. Chem. Soc., 1926 (1966). W JII'F'. ‘::: 45. 46. 47. 48. 49. 50. 51. 52. 53- 54. 56. 57- 58. 59- 60. 61. 62. 63. 64. 65. 66. . Schneider and L. K. Lang, Chem. Ccmn., 13 (1967). . v. . R. . B. . H. O H. S Interscience div. ., John Wiley 8: Sons, Inc, New York (1966). . Meerwein, Angew. Chem., _61, 374 (1955). . Meerwein, v. Hederick, H. Morschel and K. Wunderlich, Ann, 635, J A; 1 (1960). 204. . Winstein, Grunwald and Ingraham, J. Am. Chem. Soc., 19, 821 (1948). Brutcher, Jr., F. J. Vara, J. Am. Chem. Soc., 18, 5695 (1956). Grinstead, J. Org. Chem., 26, 238 (1961). Anderson and S. Winstein, J. Org. Chem. , _2_8_, 605 (1963). Olson, Tetrahedron Letters, 12, 2053 (1966). Cordes, in "Progress in Physical Organic Chemistry," Vol IV, r] C. Cohen, A. Streitwieser, Jr. and R. W. Taft, Editors, ' ; E. N. Marvell and M. J. Joncich, J. Am. Chem. Soc., 13, 973 (1951). L. A. Cort and R. G. Pearson, J. Chem. Soc., 1682 (1960). (a) A. Rieche, E. Schmitz, W. Schade and E. Beyer, Ber., 2_, 2926 (1961). (b) J. D. Prugh and W. C. McCarthy, Tetrahedron Letters, 13, 1351 (1955) s. Hunig, Angew. Chem., 1_, 400 (1964), Angew. Chem. Internet. Edit., 3; 548 (1964) N. C. Deno, H. C. Richey, Jr., N. Friedman, J. D. Hodge, J. J. Houser, and C. U. Pittman, Jr. ., J. Am. Chem. Soc. ., 85, 2991 (1963). D. M. Brouwer, Chem. Comm., 515 (1967). G. A. Olah, D. H. O'Brien and A. M. White, J. Am. Chem. Soc., 82, 5694 (1967). G. G. R. R. A. A. W. W. Olah and D. H. O'Brien, ibid., 82, 1725 (1967). Olah, J. Sommer, and E. Namanworth, ibid., 89, 3576 (1967). Hoffmann, Ang. Chemi intern. Edit., 11, 977 (1965). Hoffmann and J. Schneider, Ber., 100, 3689 (1967). M. J. S. Dewar and A. P. Marchand, J. Am. Chem. Soc., E3, 354 (1966). G. Frankel and J. P. Kim, ibid., 88, 4203 (1966). 67. 68. 69. 70. 71. 72. 73 - 74. 75. 76. 77. 78. 79- 81. 82. 83. 85. 86. J. C. . U. Pittman, Jr., and G. A. Olah, 11.18, 81, 2998, 5123 ((1965). . Deno, H. G. Richey, Jr., J. S. Liu, J. D. Hodge, J. J. Houser and . G. Richey, Jr. and J. M. Richey, ibid, 88, 4971 (1966). 205 D. Roberts and w. T. More1and, Jr., ibid., 15, 2168 (1953). .E. K. Branch and M. Calvin, "The Theory of Organic Chemistry, " Prentice-Hall, Inc. , New York, N.Y. 1941, pp 216—220. . H. Jaffe, Chem. Revs., 23, 191 (1953). Stewart and K. Yates, J. Am. Chem. Soc., 82, 4059 (1960). Hart and P. A. Law, ibid., 88, 1957 (1964). M. J. Wisotsky, ibid, 85, 2016 (1962). von R. Schleyer and G. w. Van Dine, ibid, 88, 2322 (1966). . C. Deno, et.a1., ibid, 85, 2991, 2995, 2998 (1963),- 86, 1871 (1964). . Simonetta and E. Heilbronner, Theoret. Chim. Acta. (Berlin), 2, 228 (1964). . M. Hirst and J. w. Limmett, J. Chem. Soc., 1035 (1962). L. Weber, M. W. White and K. V. Manning, "College Physics," McGraw- Hill Book Co., New York-Toronto-Iondon, 2nd Ed. , p. 385 (1952). M. Bollinger, C. A. Cupas, K. J. Friday, M. L. Woolfe and G. A. Olah, J. Am. Chem. Soc., 82, 157 (1967). . C. Brown and Y. Okamoto, ibid, 88, 4979 (1958). . V012 and M. J. Volz deLecea, Tetrahedron Letters, 38, 3413 (1965). O . Hart and D. A. Tomalia, ibid, 15, 1347 (1967). . V012 and M. J. Volz de Lecea, ibid, 32, 4683 (1966). . w. Murray and M. L. Kaplan, ibid, 111, 1307 (1967). . A. Olah and A. M. White, J. Am. Chem. Soc., 20, 1884 (1968). "(11141111mm 3