PRODUCE AND RATES OF DECOMPOSQTKON OF SOME BSCYCLBC DEACYL FEROXlDES Thesis for fire Degree of DH. D. MICHEGAN STATE UNWERSITY Frank J. Chloupek 1961 xxxxxx MSU Li‘sQl Lfinbdlfiu, INSIDHIMN RETURNING MATERIALS: PIace in book drop to remove this checkout from LIBRARIES __ your record. FINES will be charged if book is I returned after the date I stamped below. 4//2L . Arm? ABSTRACT PRODUCTS AND RATES OF DECOMPOSITION OF SOME BICYCLIC DIACYL PEROXIDES by Frank J. Chloupek The purpose of this investigation was to obtain information concerning the behavior of free radicals in the bicyclo [2, 2, 1] heptane ring system. In particular, the question of whether or not partici- pation would be significant in radical reactions was examined. Four diacyl peroxides derived from 2-exo- and 2-endonorbornanecarboxylic acids, and 5-exo- and 5-endonorbornenecarboxylic acids were pre-- pared, and their rates and products of decomposition in carbon tetrachloride studied. The principal products from Z-endonorbornanecarbonyl peroxide were carbon dioxide (79. 3%), 2-endonorbornyl 2-endonorbornane- carboxylate (10.2%), 2-endonorbornanecarboxylic acid (11. 2%) and 2-exochloronorbornane (not determined quantitatively). Z-Exonor- bornanecarbonyl peroxide produced carbon dioxide (73. 7%), 2-exonor- bornyl 2-exonorbornanecarboxylate (l4. 8%), Z-exonorbornanecarboxylic acid (6. 2%) and 2-exochloronorbornane (not determined quantitatively). 5-Endonorbornenecarbonyl peroxide yielded carbon dioxide (48. 5%), S-endonorbornenyl 5-endonorbornenecarboxylate (9. 7%), the gamma-3 lactone of Z-exotrichloromethyl-3-endohydroxy-5-endonorbornane- carboxylic acid (39%), 5-exochloronorbornane (not determined quanti- tatively), and polychloroalkanes (not determined quantitatively). 5-Exonorbornenecarbonyl peroxide produced carbon dioxide (56.4%), 5~exonorbornenyl 5-exonorbornenecarboxylate (14.6%), a lactone 2 Frank. J. Chloupek (15. 3%), an acid (not determined quantitatively), 5=exochloronor~ bornene (not determined quantitatively), and polychloroalkanes (not determined quantitatively). The production of the gamma=3~lactone of 2=exotrichloromethyl~ 3-endohydroxy-5-endonorbornanecarboxylic acid from 5—endonor- bornenecarbonyl peroxide represents a novel induced decomposition involving an intramolecular attack of a radical on a carboxyl group. The production of a lactone from 5-exonorbornenecarbonyl peroxide is evidence for the occurrence of a free radical Wagner-Meerwein rearrangement. The rates and products from decomposition produced little or no evidence for participation in free radical reactions. Steric crowdu ing in 2-endonorbornanecarbonyl peroxide appears to be significant. The relative rates of'decomposition of the various peroxides at 44. 50 in carbon tetrachloride, and their energies of activation are given below. Compound éQCODO); qu AC9 O O: QJZ )2 Relatlve 1 8 10.8 11. 2 rate E k . 8” cal 32.8 27.1 24.0 24.7 /mole) Several new compounds in the bicyclo [2, 2, l] heptane series were prepared. Some miscellaneous experiments involving the attempted preparations of 7-norbornenecarboxylic acids and the identifi- cation of the ester produced from the decomposition of cyclopropylacetyl peroxide are desc ribed. PRODUCTS AND RATES OF DECOMPOSITION OF SOME BICYCLIC DIACYL PEROXIDES BY Frank J. Chloupek A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCT OR OF PHILOSOPHY Department of Chemistry 1961 ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Professor Harold Hart for his encouragement and guidance throughout the course of this investigation. Grateful acknowledgment is extended to Dr. William H. Reusch and Dr. Morley Russell for helpful suggestions and- discussions. Appreciation is also extended to the National Science Foundation whose fellowship program provided personal financial assistance from September, 1959 through June, 1961. ************* ii To Trilby, My Wife, for Her Patience, and to My Late Father for His Faith. iii TABLE OF CONT ENTS Page INTRODUCTION AND HISTORICAL . ............ .. . 1 RESULTS AND DISCUSSION ................... 9 I. Decomposition Products from Diacyl Peroxides . . . . 9 A. Products from Saturated Peroxides ..... . . . 9 1. Products from Z-Endonorbornanecarbonyl Peroxide..... ..... ,. ..... 9 2. Products from Z-Exonorbornanecarbonyl Peroxide......_...... ..... 12 B.. Products from Unsaturated Peroxides . . . . . . . 12 1. Products from 5-Endonorbornenecarbonyl Peroxide . . . ................ 12 2. Products from 5-Exonorbornenecarbonyl Peroxide . .................. 20 11. Rates of Decomposition of Diacyl Peroxides ....... 24 EXPERIMENTAL . . . . . . . .................. 32 1. Apparatus and Reagents . . . . . . . . . . . . . . . . . 32 A. Apparatus . ..................... 32 B. Purification of Carbon Tetrachloride ...... . . 32 C. Standardization of Sodium Thiosulfate . . . . . . . 34 II. Preparation of Diacyl Peroxides. . . . . . . . ..... 34 A. Preparation of Acids ..... . . . ...... 34 1. Preparation of 5- Endonorbornenecarboxylic Acid. . _. . . ................ 34 2. Preparation of Methyl S—Endonorbornene- carboxylate. .......... ‘. . . ..... 36 3. Preparation of 5- -Exonorbornenecarboxylic Acid .............. . ........ 36 4. Preparation of 2- -Endonorbornanecarboxylic Acid........... ............ 37 5. Preparation of 2-Exonorbornanecarboxylic Acid. . . . . . ............... 38 B. Preparation of Acid Chlorides . . ......... 38 C. Preparation of Diacyl Peroxides .......... 41 iv TABLE OF CONTENTS - Continued Page III. Preparation of Lactones .................. 48 A. Preparation of the Gamma-3 Lactone of 2-Exo- bromo-3-endohydroxy-5-endonorbornanecarboxylic Acid .............. ...........48 B. Preparation of the Gamma-3 Lactone of Z-Exochloro- 3-endohydroxy-S-endonorbornanecarboxylic Acid . 48 C. Preparation of the Gamma-3 Lactone of Z-Exoiodo- 3-endohydroxy-5-endonorbornanecarboxylic Acid . 50 IV. Preparation of Esters .................. . 50 A. Preparation of Alcohols . . . ............ 50 1. Preparation of 5-Endohydroxynorbornene . . . . 50 2. Preparation of 5-Exohydroxynorbornene . .h . . 53 3. Preparation of 2-Endoacetoxynorbornane . . . . 53 4. Preparation of 2- -Endohydroxynorbornan‘e . . . . 56 5. Preparation of 2- -Exohydroxynorbornane . . . 56 6. Preparation of 5- Endohydroxymethylnorbornene 57 7. Preparation of 5- -Exohydroxymethylnorbornene. 57 B. Preparation of Esters ............. . . . 59 V. Decomposition of Diacyl Peroxides .......... .’ . 66 A. Identification of Products from Decomposition . . . 66 1. 2-Endonorbornanecarbonyl Peroxide . '. . . . . 66 2. 2-Exonorbornanecarbonyl Peroxide ...... '. 71 3. 5-Endonorbornenecarbonyl Peroxide ...... 71 4. 5-Exonorbornenecarbony1 Peroxide Products . . 84 B. Determination of Products from Decomposition . . . 91 1. Determination of Carbon Dioxide . . . . . . . . 91 2. Determination of Carbonyl Containing Products. 93 C. Kinetics of Decomposition ........... . . . 93 VI.. Miscellaneous Experiments ....... . ..... . . . 95 A. Preparation of'Galvinoxyl . . .‘ . . ......... 95 B. Determination of Purity of Norbornenyl Acids. . . . 95 C. Attempted Preparation of the Gamma-3 Lactone of 2-Exotrichloromethyl- 3 - endohydroxy- 5- endonor- bornanecarboxylic Acid ............... 95 D. Attempted Preparation of Syn- and Anti-7-nor- bornenecarboxylic Acids .............. 96 1. Preparation of 7-Syn-bromonorbornene ..... 96 2. Preparation of Syn- and Anti-7-carbomethoxy- norbornene .............. . . . . . 97 3. Preparation of Acetyl Bromide ......... 97 TABLE OF CONTENTS - Continued Page . Acylation of Norbornene ............. 98 . Haloform Oxidation of Acylation Product . . . . 98 Dehydrohalogenation of Oxidation Product . . . 100 . Preparation of n- Butyl Lithium ......... 102 Preparation of Cyclopentadienyl Lithium . . . . 102 Preparation of the Exocyclic Enol Acetate of Acetylcyclopentadiene ........... . . 102 10. Reaction of the Exocyclic Enol Acetate of Acetylcyclopentadiene with Ethylene . . . . . . 103 \OCDKICFU'Irk 11. Acylation of Norbornene with Phosgene ..... 103 E. Preparation of Cyclopropaneacetyl Peroxide . . . . 104 1. Preparation of Cyclopropyl Lithium ....... 105 2. Preparation of B-Cyclopropylethanol . . . . . . 105 3. Preparation of Cyclopropaneacetic Acid ..... 105 4. Preparation of Cyclopropaneacetyl Chloride . . 106 5. Preparation of Cyclopropaneacetyl Peroxide . . 106 F. Preparation of Cyclopropylcarbinyl Cyclopropane- acetate ........................ 107 . Preparation of Cyclobutanol ............ . 107 . Preparation of Cyclobutyl Cyclopropaneacetate . . . 107 Determination of the Ester Fragment from the Decomposition of Cyclopropylacetyl Peroxide . . . 109 5510 SUMMARY ......... . ................... 112 LITERATURE CITED ................... . . . . 114 APPENDIX .......................... . . . 118 vi TABLE 10 Products Peroxide . Products Peroxide . Products Peroxide . Products Peroxide LIST OF TABLES Page of Decomposition of Z-Endonorbornanecarbon l in Carbon Tetrachloride after 72 Hours at 78 . 10 of Decomposition of 2- Exonorbornanecarbonyl in Carbon Tetrachloride after 72 Hours at 780 13 of Decomposition of 5-Endonorbornenecarbon l in Carbon Tetrachloride after 72 Hours at 78 . 14 of Decomposition of 5- Exonorbornenecarbonyl in Carbon Tetrachloride after 72 Hours at 780 21 Rate Constants and Activation Parameters for the Decomposition of Various Peroxides in Carbon Tetra- chloride Determined by Titrimetric Techniques . . . . . 26 . Rate Constants and Activation Parameters for the Decomposition of Various Peroxides in Carbon Tetra- chloride Determined by Infrared Techniques ....... 27 Rate Constants and Activation Parameters for the ‘Appearance of Ester from the Decomposition of Various Peroxides in Carbon Tetrachloride Determined by Infra- redTechniques.............. ..... 28 . Rate Constants and Activation Parameters for the Appearance of Acid from the Decomposition of Various Peroxides in Carbon Tetrachloride Determined by Infra- redTechniques..... ...... 29 ilodate as 10. 11. . Standardization of Sodium Thiosulfate Using Potassium aPrimaryStandard.............. 35 Standardization of Sodium Thiosulfate Using Benzoyl Peroxide as a Primary Standard. . . . . . ....... 35 Yields and Physical Properties of Various Acid Chlorides................... ...... 42 vii LIST OF TABLES - Continued TABLE 12. 13. 14. 15. 16. 17.- 18. 19. 20. 21 Yields and Physical Properties of Various Diacyl Peroxides .................. . . . Yields and Physical Properties of Various Esters. . . . Decomposition of 2- -Exonorbornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 90 Determined by Titri- metric Techniques . . ................ Guggenheim Data for the Decomposition of 2-Exonor- bornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 9 Determined by Titrimetric Techniques ...... Decomposition of 2- -Exonorbornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 9o Determined by Infrared Techniques ...... . ........ . . . . . . . . . Guggenheim Data for the Decomposition of 2- Exonor- bornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 9o Determined by Infrared Techniques . . q ...... .Appearance of Acid from the Decomposition of 2- Exo- norbornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 9o Determined by Infrared Techniques . ..... Guggenheim Data for the Appearance of Acid from the Decomposition of 2- -Exonorbornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 90 Determined by Infra- red Techniques . . . ~ .................. .. .Energy of Activation for the Decomposition of 2-Exo- norbornanecarbonyl Peroxide in Carbon Tetrachloride Determined by Titrimetric Techniques . . . ., ..... .. EntrOpy of Activation for the Decomposition of 2-Exo- norbornanecarbonyl Peroxide in Carbon Tetrachloride Determined by Titrimetric Techniques . ........ viii Page 43 61 121 121 124 124 129 129 132 132 LIST OF FIGURES FIGURE Page 1.. Diagram of the Apparatus Used for Carbon Dioxide Determination . ....... . . . . . . . . . . ..... 33 2. Infrared Spectrum of 2-Endonorbornanecarboxylic Acid . 39 3. Infrared Spectrum of 2-Exonorbornanecarboxylic Acid 40 4-1nfrared-Spectrum of 5-Endonorbornenecarbony1 Peroxide .- . ....... . ................ 44 5. Infrared Spectrum of 5-Exonorbornenecarbonyl Peroxide 45 6. Infrared Spectrum of 2-Endonorbornanecarbonyl PerOJdde o o ooooo o o o o o o o o‘ o o ooooooooo 46 7. Infrared Spectrum of 2-Exonorbornanecarbonyl Peroxide 47 8. Infrared Spectrum of the Gamma-3 Lactone of 2-Exo- bromo-3-endohydroxy-S-endonorbornanecarboxylic Acid 49 9. Infrared Spectrum of the Gamma-3 Lactone of 2-Exo- chloro-3-endohydroxy-5-endonorbornanecarboxylic Acid 51 10. 11. 12. 13. 14- Infrared Spectrum of the Gamma-3 Lactone of 2-Exoiodo- 3-endohydroxy-S-endonorbornanecarboxylic Acid . . . . . Infrared Spectrum of 5-Endonorbornenyl 3, 5-dinitro- benzoate..... ........ ....... Infrared Spectrum of 5-Exonorbornenyl 3, 5-dinitro- benzoate. . . . ..... . ................. Infrared Spectrum of the 3, 5-Dinitrobenzoate of 5-Endo- hydroxymethylnorbornene .............. . . Infrared Spectrum of the 3, 5-Dinitrobenzoate of S-Exo- hydroxymethylnorbornene ................. ix 52 54 55 58 60 LIST OF FIGURES - Continued FIGURE 15. 16. 17. 18.- 19. 20. 21.- 22. 23. 24. 25. 26. 27. Infrared Spectrum of 5-Endonorborneny1 5-Endonor- bornenecarboxylate .................... Infrared Spectrum of 5-Exonorbornenyl 5-Exonor- bornenecarboxylate ............ . ....... Infrared Spectrum of Z—Endonorbornyl 2-Endonor- bornanecarboxylate o o o o ...... ' o o o o o o o o o 0 Infrared Spectrum of Z-Exonorbornyl Z-Exonorbornane- carboxylate . ..... . .............. Infrared Spectrum of the Alkyl 'Halide obtained from-the Decomposition of 2-Endonorb6rnanecarbonyl Peroxide . . Infrared Spectra of 2-Exochloronorbornane and Z—Endo- chloronorbornane ...... ‘. . . ....... . . . . . Infrared Spectrum of the Ester obtained from the Decomposition of 2-Endonorbornanecarbonyl Peroxide . Infrared Spectrum of the Acid .obtained from the Decomposition of Z-Endonorbornanecarbonyl Peroxide . Infrared Spectrum of the Alkyl Halide obtained from the Decomposition of 2-Exonorbornanecarbonyl Peroxide . .. Infrared Spectrum of the Ester Obtained from the Decomposition of 2-Exonorbornanecarbonyl Peroxide . .. Infrared Spectrum of the Acid obtained from the ' Decomposition of 2-Exonorbornanecarbonyl Peroxide . . Infrared Spectrum of the Alkyl Halide obtained from the Decomposition of 5-Endonorbornenecarbonyl Peroxide . Infrared Spectra of 5-Exochloronorbornene, 5-Endo- chloronorbornene, and 3-Chloronortricyclene ...... Page 62 63 64 65 . 67 68 69 70 72 73 74 75 76 ~LIST OF FIGURES - Continued FIGURE 28.. 29.- 30. 31.’ 32. 33.. 34. 35. 36. 37. Infrared Spectrum of the Polychloroalkane obtained from the Decomposition of 5-Endonorbornenecarbonyl ’ Peroxide.............. ..... Infrared Spectrum of the 3, 5-Dinitrobenzoate of the Alcohol obtained from the Decomposition of 5-Endonor- bornenecarbonyl Peroxide (Distillate) ...... ‘ . . . . . InfraredSpectrum of the 3, 5-Dinitrobenzoate of the Alcohol obtained from the Decomposition of 5-Endonor- bornenecarbonyl Peroxide (Pot Residue) . . . ...... Infrared Spectrum of the Lactone obtained from the Decomposition of 5-Endonorbornenecarbonyl Peroxide . Nuclear Magnetic Resonance Spectra of Lactones . . . . Infrared Spectrum of the Acid obtained from the Hydrolysis of the Lactone obtainedfrom the Decompo- sition of 5-Endonorbornenecarbonyl Peroxide ...... Infrared Spectrum of the Alkyl Halide obtained from the Decomposition of 5-Exonorbornenecarbonyl Peroxide . . Infrared Spectrum of the Polychloroalkane obtained from the Decomposition of 5-Exonorbornenecarbonyl Peroxide Infrared Spectrum of the 3, 5-Dinitrobenzoate of the Alcohol obtained from the Decomposition of 5-Exonor- bornenecarbonyl Peroxide (Distillate) . . . . . . . . . . Infrared Spectrum of the 3, 5-Dinitrobenzoate of the , Alcohol obtained from the Decomposition of 5-Exonor- 38.- 39. bornenecarbonyl Peroxide (Pot Residue) . ........ Infrared Spectrum of the Acid obtained from the Decom- position of 5-Exonorbornenecarbonyl Peroxide . . Page 78 79 8O 82 83 85 87 88 89 90 92 Infrared Spectrum of the Bromoketone obtained from the ~ Acylation of Norbornene with Acetyl Bromide ..... xi 99 LIST OF FIGURES - Continued FIGURE 40. 41 42 43. 44. 45. 46. 47. 48.. 49 50. , Infrared Spectrum of the Bromoacid obtained from the Haloform Oxidation of the Bromoketone obtained from the Acylation of Norbornene with Acetyl Bromide . . . . . Infrared Spectrum of Cyclopropylcarbinyl Cyclopropane- ac etate ........... O O OOOOOOOOOOOOOO .- Infrared Spectrum of Cyclobutyl Cyc10propaneacetate . . Infrared Spectrum of the Ester obtained from the Decomposition of Cyclopropaneacetyl Peroxide . . . Titer Versus Time Plot for the Decomposition of 2-Exo- norbornanecarbonyl Peroxide at 53. 9 ..... Guggenheim Plot for the Decomposition of 2- Exonor- bornanecarbonyl Peroxide at 53. 90 (Titration). . Quantitative Infrared Spectra of the Decomposition of 2- -Exonorbornanecarbony1 Peroxide at 53. 9o ..... Absorbancy Versus Time Plot for the Decomposition of 2-Exonorbornanecarbonyl Peroxide at 53. 9 ...... Guggenheim Plot for the Decomposition of 2- Exonor- bornanecarbonyl Peroxide at 53. 90 (Infrared) ...... .~ Absorbancy Versus Time Plot for the Appearance of Acid from the Decomposition of 2- Exonorbornane- carbonyl Peroxide at 53. 9O .............. Guggenheim Plot for the Appearance of Acid from the Decomposition of 2- ~Exonorborn'anecarbonyl Peroxide at53.9(Infr.ared). xii Page 101 108 110 111 122 123 125 127 128 130 131 INTRODUCTION AND HISTORICAL The bicyclo [2, 2, 1] heptane ring system has been a model structure in the study of reaction mechanisms, particularly because of its well-defined geometry and its propensity for molecular rearrange- ment. The classical Wagner-Meerwein rearrangement and a major part of nonclassical carbonium ion theory have been developed using this ring system (1, 2, 3). Although this thesis is concerned with free radical reactions in the bicycloheptane series, a brief review of pertinent ionic reactions will be helpful. Roberts (4) found that the solvolysis of.2-exochloronorbornane in 80% aqueous ethanol was 70 times faster than the corresponding endo isomer. In the exo case, the ethylene bridge can assist the ionization since it is trans to the departing chlorine (equation 1). In 2-endochloro- norbornane, the ethylene bridge lies cis to the-departing chlorine and the ionization must proceed without assistance (equation 2).' 55 ""‘> H20 A m 1735* ' 1 H . HO : c1' “—1" % W Q0}, ‘2’ O In the endo case, once the chlorine is ionized, the delocalization of appropriate bonds may occur to give the same intermediate as in the exo case. This would account for the production of only exo alcohol from either. isomer. An alternate but perhaps less satisfactory explanation for the difference in rates may be steric hindrance to ionization. In the endo chloride, the change in bond hybridization from sp3 to the Sp?- hybridi- zation of the carbonium ion would force the departing chlorine further under the ring in close proximity to the endo hydrogens. By using C“ labeling, Roberts (5) found considerable scrambling in. the product, indicating hydride shifts as well as skeletal rearrange- ments . W:.:®:® Roberts (4) also found that the solvolysis of 5-exochloronor- bornene in 80% aqueous ethanol was 150 times faster than the corres- ponding endo isomer. This he attributed to the ability of the double bond to participate in the ionization of the chloride. Again the exo isomer has the necessary trans configuration. Alkyl rearrangements of free radicals are not as well known. Berson (6) has presented evidence for a free radical WagneraMeerwein rearrangement in the decomposition of 2, 2',-bisazoca-mphane. Along with camphane and the disproportionation products, Berson obtained 2,3,3—trimethyl-4-ethylcyclopentene and isocamphane. The same products were formed in the decomposition of the azo compound prepared from campholenaldehyde. . Berson postulates the following __ Nj I 1 1" J7 The work described in this thesis was undertaken to examine the 14% I/ N Ste 11 behavior of free radicals in the bicyclo[2, 2, 1]heptane ring system. In particular, the question of whether or not participation would be significant in radical reactions, as it is known to be in ionic reactions, was examined. . Four diacyl peroxides, derived from 2-exo—and 2- endonorbornanecarboxylic acids, and 5-exo- and 5-endonorbornene- carboxylic acids were prepared, and their rates and products of decomposition in carbon tetrachloride studied. . Diacyl peroxides are of considerable interest in that they provide a convenient route to free radicals; are generally easy to prepare, and are sufficiently unstable to allow investigation not too far above room temperature. A number of textbooks (9, 10, ll, 12) contain excellent reviews of diacyl peroxide decomposition. There are essentially two modes of diacyl peroxide decompo- sition, the spontaneous fission of the oxygen-oxygen bond homolytically to form radicals, O O O II II . ll (6) RCOOCR ——-—9 ZRCO- or heterolytically to form ions, 0 O O O II H II H G RCOOCR -—-> RCO® + RCO (7) and the corresponding induced decomposition by radicals (13, 14) or by ions (15). The decomposition of a dilute solution of a symmetrical diacyl peroxide in a non-polar solvent provides the most favorable conditions for homolytic fission. This appears to be a direct result of the small bond dissociation energy of the oxygen-oxygen linkage. This energy has been estimated at 27-35 kca1./mole (16). . A major problem in the homolytic mode of decomposition is the extent of carbon-carbon breakage in the rate-determining step. The decomposition can proceed in any or all of the three paths shown in equations 8, 9, and 10. f? f? — u f? ‘1 . l .RCOOCR ——) LRCO """" OCR —-———> 2RCO. (8) fi ‘11 I O fi 1? ll ° RCOOCR -——-> R ----- C—O° ' ' 'OCR -—-9- R- + C0; + °OCR (9) 0 R R P. H RCOOCR-——->‘ R'°°°C-O---O—C-°°.'R --—->2R°+2CO,_ (10) When R is aryl, there is good evidence for equation 8 as a model for Spontaneous decomposition. . Hammond and Soffer (17) obtained benzoic acid quantitatively by decomposing benzoyl peroxide in the presence of iodine and water. ' In the absence of water, iodobenzene was obtained in better than 90% yield.) 0. o‘0 :r: _ O— O 2 +2HI H20 .1 .1 9 ‘CO- 2 O (30- I2 2 0 C01 / Z \ 2 Q + 2co2 Q (11) Ford and Mackay (18, 19) and also Teller (20) in their work with heterocyclic diacyl peroxides have shown that even in the absence of inhibitors, only small amounts of carbon dioxide are liberated and the major product is the corresponding acid. [I ' ng' _ -H .> 1 IL COZH (12) s 2 S In the presence of halogenated benzenes, the phenyl ester and halogen are major products. Walling and Hodgdon (7) reported that in the decomposition of acetyl peroxide in the presence of iodine and water, no acetic acid is produced, the main product being methyl iodide. This supports equation 10 but Szwarc (21) and also DeTar and Weis (22) have sub» stantial evidence to indicate the absence of acid is due to the very rapid loss of carbon dioxide from the acyloxy radical. - More recently, Shine and Hoffman (8) have provided evidence for the existence of the acetoxy free radical. There is some evidence for multiple bond scission. This can be important when a radical of high stability such as t-butyl or tri- chloromethyl is produced. A notable example stems from the work of Bartlett and Leffler (23) with phenylacetyl peroxide. They found that phenylacetyl peroxide decomposes faster at 00 than benzoyl peroxide at 800, and attributed the rate increase to multiple scission as shown in equation 9. Homolytic fission is often accompanied by induced decomposition due to the presence of radicals in solution. 1 Since this complicates the kinetics by giving rise to higher reaction order, it must be corrected for or eliminated. Kinetic analysis (14) or the use of a radical trap (17) has been effective- Concentration is a very important factor in controlling induced decomposition (24). The nature of the solvent (14) and temperature (24) are also factors. The position of attack of the inducing agent varies, and may be determined by isolation of the products (25,26). . Heterolytic fission of diacyl peroxides is favored by a decrease in symmetry of the molecule and the use of solvents of high polarity. The transition, state can vary from a slight polarization of the oxygen- oxygen bond to complete dissociation into ions. ‘ Leffler (15) in his study of 4¥methoxy-4'-nitrobenzoyl peroxide found that in non-polar solvents, this peroxide decomposes at the same rate as benzoyl peroxide. But in polar solvents, the rate is increased markedly. Leffler attributed this to anionic mechanism (equation 7) supplanting the homolytic fission. This is supported by the ability to induce decomposition by acids at a rate proportional to the acidity constant of the acid. » In later work,- Leffler and Petropoulos (27) found that 3, 5-dinitro-4'-methoxy- benzoyl peroxide was even more susceptible to acid than the 4-methoxy-- 18, Denny has Shown 4'-nitrobenzoyl peroxide. —Making use of oxygen (.28) that the 4-methoxy-4'-nitrobenzoyl peroxide decomposition »mechanism is not adequately represented by equation 7 since no equili- bration of oxygens was found. ' If the ions were free, the oxygens should equilibrate. Denny's postulated mechanism is shown in equation 14. 9189 (14) c11,o-.—oc OC‘NOZ Bartlett and Greene (29) in their investigation of ditriptoyl peroxide, have made a detailed product study which indicates the operation of'three separate mechanisms in the decomposition. CO; (50%), triptycene (45%), -H free acid (15%), combined acid (15%) (15) hydroxytriptycene (17%) ,' ester (6%) In the presence of iodine, 45% of iodotriptycene was obtained, but the amount of alcohol remained 17%. Equation 9 accounts for the major produCts while equations 8 and 14 account for the lesser products. The formation of,ester will be discussed in the following paragraph. The formation of ester in diacyl peroxide decompositions can run from 0-100% of theoretical. The mechanism of ester formation is thought to be of a cyclic intramolecular nature (22) or a geminate recombination process (30, 31). 0 '9 //O x R\ Rc'z'oo'CR ——> R — c (c: ——+ R'c'on + co,2 (16) \’\ / 0—0 9 9 § 1 RCOOCR --> R o- + 111 +002 —--> RCOR + co2 (17) Either reaction would occur in a solvent "cage. " The formation of nearly quantitative amounts of ester as found by Hart and Lau (32) in the decomposition of trans-4-t-butylcyclohexanecarbonyl peroxide tends to refute the geminate recombination mechanism since one might expect some diffusion into the solvent. There are also numerous examples of optically active diacyl peroxides that give rise to optically active esters (33, 34, 35). Although there is some racemization, this is also evidence against equation 17 since if one estimates the energy of activation for racemization at 2 kcal. /mole, one would expect random rotation of the planar free radical to occur before recombination. One must, however, take into account the fact that the reaction takes place in a solvent "cage, " and even if one considers the alkyl radical in equation 17 to be planar, a fully racemic product would necessitate o , . . a 180 rotation before recomb1nat1on. 13. ‘\ Rbo. lc _. Rco- ;c\ (18) R Thus the formation of ester may very well be a highly stereospecific recombination of radicals. RESULTS AND DISCUSSION I. Decomposition Products from Diacyl Peroxides A. Products from Saturated Peroxides 1. Products from 2-Endonorbornanecarbonyl Peroxide The products from the decomposition of 2-endonorbornanecarbonyl peroxide are consistent with other aliphatic diacyl peroxides. The major products (see Table 1) were alkyl halide, carbon dioxide, acid, and ester. The large yields of carbOn dioxide (73. 7%) and small yields of acid (1 l. 2%) and ester (14. 8%) parallel the products (of decomposition of delta-phenyl- valeryl peroxide. 1DeTar and Weis (22) found that in carbon tetrachloride, this peroxide produced 84% carbon dioxide, 4% delta-phenylvaleric acid, and 17. 5% gamma-phenylbutyryl delta-phenylvalerate. . They concluded that two acyloxy radicals were formed initially, and that subsequent loss of carbon dioxide occurred rapidly. The ester formed from 2-endonorbornanecarbonyl peroxide was 2-endonorbornyl 2-endonorbornanecarboxylate. Starting with 6. 25 mmoles of peroxide, 0. 512 mmoles of pure ester was isolated. Using infrared analysis, the total ester produced was 0. 204 mmoles/mmoles of peroxide. The similarity of infrared spectra of endo and exo isomers, and the complexity of the product spectrum did not allow setting a limit of detection on any one isomer. . On the basis of isolation, 46% of the ester formed retains the endo configuration. It must be emphasized, however, that recovery involves a sublimation and a crystallization to yield analytically pure material and handling losses are considerable. Alkyl halide was distilled along with the last traces of carbon tetra- chloride. The alkyl halide was identified as 2-exochloronorbornane by infrared but was not determined quantitatively. 9 10 Table 1. Products of Decomposition of 2-Endonorborngnecarbonyl I Peroxide in Carbon Tetrachloride After 72 Hours at 78 (1. 17 mmoles of peroxide in 50 m1. of carbon tetrachloride). Mmoles/mmoles Av. mmoles/ Product Mmoles peroxide mmoles peroxide %COZ %R 1.73 1.475 . i 1 1.473 73.7 0 . . 0 Carbon 1 72 1 47 dioxide 1 77* 1 509 b . * . 76 1.503 1.506 75.3 0 0.35 0.299 a a . Ester 0.34 0.291 0.295 14.8 ‘ 29.5 (6:489 1./g. * -cm) 0.27 0.231 b b 2:: 0.28 0.239 0.235 11.8 23.5 0.14 0.120 a a 0.124 11.2 11.2 Acid 0.15 0.128 = O . , :1: k Sing/g 0.19,, 0.162 0 162 14 b 14 8b 0.19 0.162 ' 18 .. 9,: 0. 2' M Styrene present a Accounts for 100.7%of carbonyl, 40. 7% of alkyl. b Accounts for 101.9% of carbonyl, 38. 3% of alkyl. 2-Exochloronorbornane was identified but not determined quantitatively. 11 Extraction of the decomposition products with 5% sodium carbonate produced 2-endonorbornanecarboxylic acid. Identification of the acid and the ester was based on melting points, mixed melting points, and infrared spectra. On the basis of products formed, .a mechanism can be written: (17) (:02 l + cc1, 33-» -c-c1 + CC13° 2c013- 3‘49 C2016 12 * Applying steady state approximations, there is obtained the same form ofvrate expression as calculated by Bartlett and Nozaki for the decomposition of benzoyl peroxide (14). 3 dp _ 2' For the mechanism written, the ki term is zero since no induced decomposition is illustrated. 2. Products from 2-Exonorbornanecarbonyl Peroxide. The products from the decomposition of 2-exonorbornanecarbonyl peroxide parallel those obtained from the endo isomer. The products are summarized in Table 2. The methods of detection and identification were identical to those employed in the endo case. The major products included carbon dioxide (79. 3%), 2-exonor- bornanecarboxylic acid .( 6. 2%), and 2-exonorbornyl Z-exonorbornane- carboxylate (10. 2%). The alkyl halide was identified as 2-exochloronor- bornene by infrared but was not determined quantitatively. The mechanism for this decomposition would be identical to sequence 17, and the kinetic expression identical to equation 18. B. Products from Unsaturated Peroxides 1. Products from 5-Endonorbornenecarbonyl Peroxide 5-Endonorbornenecarbonyl peroxide produced 48. 5% of the theoretical carbon dioxide. This is less than one would expect from an alkyl diacyl peroxide. .For example, Wyman (47) found 93. 5% carbon dioxide evolution from cyclohexanecarbonyl peroxide. .Table 3 summarizes the products from the decomposition of 5-endonorbornenecarbonyl peroxide. 13 Table 2. o Peroxide in Carbon Tetrachloride After 72 Hours at 78 peroxide in 50 m1. of carbon tetrachloride). Products of Decomposition of Z-Exonorbornanecarbonyl (1..11.mmoles of M Mmoles/mmoles Av. mmoles/ Product Mmoles peroxide mmoles peroxide %COZ %R 1.75 1.590 a 1.586 79.3 0 Carbon 1. 74 1. 582 . . * dwmde 1.65* 1.499 1 500 75 0b 0 1 66 1.509 ' ' 0.22 0.200 a a - Ester 0.23 0.208 0°204 10°2 20°4 (6:470 l./g. ‘* 0. . -cm.) 0 :21. g :2: 0.254 12.710 25.41D 0.24 0.218 a a Acid 0.25 0.227 0.223 6.2 6.2 (6:421:45 0.33”“ 0.300 0 2% 8 1b 8, 1b ' 0.32" 0.292 ' ' ‘ ' 9,: 0. 2 M Styrene present a Accounts for .95. 7% of carbonyl, 26. 6% of alkyl. b Accounts for '95 .8% of carbonyl, 33.5% of alkyl. 2-Exochloronorbornane was identified but not determined quantitatively. 14 Table 3. Products of Decomposition of 5-Endonorborngnecarbonyl Peroxide in Carbon Tetrachloride After 72 Hours at 78 (1. 25 mmoles of peroxide in 50 m1. of carbon tetrachloride). ‘Mmoles/mmoles Av. mmoles/ Product Mmoles peroxide mmoles peroxide %COZ %R ‘ 0. :3 0 3:: 0.969 48.5a 0 Carbon . ' dioxide >1: 1.35 1.082 b :1: 1.33 1.062 1.072 53.6 0 0.23 0.184 a. a 0.19 0.152 0.194 9.7 19.4 Ester 0. 31 0. 247 :: ’1‘ f (e 4090;“? 0.34* 0.272 b b . 0.33,“ 0.264 0.275 13.8 27.6 0.36 0.288 1.00 0.800 a 0.94 0.753 0.779 39.0 39.0a Lactone 0 . 98 0 . 784 = :1: (6 75:11:45 0.71,, 0.568 b b ° 0.62* 0.503 0.522 26.1 26.1 0.63 0.496 >1: 0. 2‘ M. Styrene present a Accounts for 97. 2% of carbonyl, 58.4% of alkyl. Accounts for 93. 5% of carbonyl, 53. 7% of alkyl. 5-Exochloronorbornene was identified but not determined quantitatively. 15 5-Endonorbornenyl 5-endonorbornenecarboxylate was produced to the extent of 19.4%. The ester was identified by extracting the crude products with base, treating the basic and/or neutral fraction with lithium aluminum hydride, and derivatizing the resulting alcohols. The only crystalline material isolated retained the endo configuration. . The com- plexity of the infrared Spectrum of the products, and the great similarity of the infrared Spectra of compounds having either exo or endo configura- tions made it impossible to set a minimal limit of detection on the amount of endo-endo isomer. Alkyl halide codistilled when the last traces of carbon tetrachloride were removed from the reaction mixture. Separation using a Beckrnan Megachrom vapor phase chromatograph at 1500 produced only trace amounts of 5-exochloronorbornene. . This alkyl halide was identified by its infrared _spectrum. When the distillation of the reaction mixture was continued, material containinga carbonyl absorbtion at 5. 58 p. in the infrared sublimed. The material was crystallized from ether and melted at 100-1010. Elemental analysis indicated an empirical formula C8H9OZC13. The chlorines were extremely resistant to hydrolysis which indicates a trichloromethyl group (54). The infrared carbonyl absorbtion indicated a lactonic material and comparison with known lactones in the norbornane ring system bears out this hypothesis. Comparison of the n.m. r. spectrum of this compound with those of known lactones also indicates a lactonic structure. . This compound is most likely the gamma-#3-lactone of 2-exotrichloro- methyl- 3 - endohyd roxy- 5 - endonorbornanec arboxylic ac id. ‘ (A) 16 It can arise by attack of a trichloromethyl radical on one of the double bonds of the peroxide molecule with the resultant radical attacking either the carbonyl oxygen or the peroxidic oxygen with the displacement of an acyloxy radical. Perhaps a more likely second step would be the dis- placement of carbon dioxide and an alkyl radical. ' CCl . CC13° ‘1‘ CO)» . o 2 // c-o-o-c / \ O o/ o + cc ( 13 c -* Do (19) // o ‘ OéC-O-O-CQO + + coz I One cannot disregard the fact that one end of the double bond is as susceptible to attack as the other. This would lead to the addition of the elements of carbon tetrachloride across (the double bond with subsequent loss of carbon dioxide and abstraction of another chlorine atom. - Lactone formation is not as likely in this case. The distance of the carboxyl group from the radical site is much greater. 17 ~'CC13 ' W‘ 9 C13C /m ...O- 0. ..Oc\\ /-o- o C\\o 20 fl? —> + CCl . O5:- oeo- -c\O 0130 3 (20) :lfi/C WW C1 I C130 + --O O- C\\O C13C . + 2C02 . O 0 C1 C1 + cm, ———> me c1 + cc1,‘. c130 3 Polychlorinated alkanes were isolated from the reaction mixture but they Could not be purified sufficiently for positive identification. A typical Sample analyzed for 60% chlorine. ‘ This would also account for the Production of only trace amounts of alkyl halide since the latter could be attaCked in- a similar manner. - Some of the ester probably had the elements of carbon tetrachloride added also. - Benzotrichloride, tetrabromoethane, bromotrichloromethane, and iodine in carbon tetrachloride were used as solvents to test the (effect of sol-Vent on lactone formation. . The amount of lactone remained virtually Constant (39% in the case of carbon tetrachloride). -In the use of iodine, t he 1actone may have been iodolactone (identification only by infrared). 18 Since this mode of decomposition is of an induced nature, inhibitors should reduce the amount of lactone formed as well as retard the rate of decomposition. -A four-fold excess of styrene produced only a 25% decrease in the amount of lactone formed and galvinoxyl produced no noticeable effect. This would indicate that the norbornene double bond is about twelve times more reactive toward a CC13 radical than the styrene double bond. It is conceivable, however, to have attack on the double bond by other radicals (e. g. styryl radical) giving rise to complex lactones. The presence of a four-fold excess of norbornene produced no noticeable change in the rate or in the amount of lactone formed. Presumably the norbornene simply acts as a chain transfer agent. Cl3C + CCl3' -.-——> (21) C130 C13C +. 0014 ——-> +. col,- Cl Kharasch (66) found‘that the elements of bromotrichloromethane could beadded to a double bond using light or organic peroxide to initiate the reaction. - Independent synthesis of the postulated trichloromethyl lactone was attempted using 5-endonorbornenecarboxylic acid or its methyl ester and either U. V. light or a trace of benzoyl peroxide. Bromotrichloromethane was used as the solvent. These methods proved unsuccessful. The chlorines of the trichloromethyl group-could not be s.olvolyzed using silver nitrate in aqueous acetone. Fuming nitric acid produced no reaction but concentrated sulfuric acid (54) did give some hydrolysis product. 19 The acid obtained analyzed for C9H1004 and had carbonyl absorbtions in the‘infrared at 5.58 (J. and 5.87 (J. (figure 33). C13C ‘ H02 0 :0 (22) The following mechanism is postulated. fl. __,., . O O// )-2 1 k2 /‘ —-> + CO, C // " 0' O (23) ZCC13° % C2C16 20 Reactions involving the addition of carbon tetrachloride to'the double bond of the alkyl halide, ester, and equations such as 20 have been omitted since they are chain transfer processes and will not appear in the rate expression when steady state approximations are applied. A steady state treatment of the above equations leads to equation 18 where ki=k4“jk1/k6 - 2- Products from 5-ExonorbornenecarbonyllPeroxide As in the case of the endo isomer, the production of carbon dioxide was low (56.4%). The yield of 5-exonorbornenyl 5-exonorbornenecarboxylate amounted to 29. 1%. The ester was identified by a degradative procedure identical torthe one described for 5-endonorbornenyl 5-endonorbornene- carboxylate. The complexity of the infrared spectrum of the products and similarity of spectra of compounds having either exo or endo configurations prevented the establishment of minimal limits on the amount of exo-exo ester produced. The only crystalline derivatives isolated retained the exo configuration. ~ Only trace amounts of alkyl halide‘could be obtained. The alkyl halide was distilled along with the last traces of carbon tetrachloride“. Separation on a BeckrnannMegachrom vapor phase chromatograph at 150CD produced the alkyl halide. Its retention time and infrared spectrum were identical to those of the alkyl halide obtained from the endo peroxide and it was identified as 5-exochloronorbornene by its infrared spectrum; The data * are summarized in Table 4. ’When the decomposition products were extracted with base, an acid was isolated from the basic extracts. Elemental analysis indicated an empirical formula C9HmOzCl4 and structures B have been assigned to this acid. 01,0 01 and/or ~ COZH cozn (B) (:1 Cl3C 21 Table 4. < Products of Decomposition of 5-Exonorborne13ecarbonyl Peroxide in Carbon Tetrachloride After 72 Hours at 78 (1. 25 mmoles of peroxide in 50 ml. of carbon tetrachloride). Mmoles/mmoles Av. mmoles/ Product ~Mmoles peroxide mmoles peroxide ’%COZ %R 1. 39 1.110 b . 1.128 56.4 0 Carbon 1.43 1.145 dioxide * ‘C 1'52* 1'215 1.215 60.8 0 1.52 1.215 1 0.37 0.296 b b 0.36 0.288 0.291 14.6 .29.1 Ester 0.37 0.296 6:4: . . ' * < 3:131? 0.53,, 0.424 ' 0.51 0.408 0.403 20.1“ 40.3C 0.47* 0.376 0.38 0.303 b b 0.38 0.303 0.306 15.3 15.3 Lactonea 0. 39 0. 312 e: . . * ( 76:30? 0.41* 0.328 C ' 0.42* 0.336 0.333 16.7 16.7C 0.42 0.336 4: 0. 2M Styrene present a e is estimated to be the same as in 5-endonorbornenecarbonyl peroxide. b . Accounts for 86. 3% of carbonyl, 44.4% of alkyl. c Accounts for 97.6% of carbonyl, 57. 0% of alkyl. 5-Exochloronorbornene was identified but not determined quantitatively. 22 As in the endo case, a great amount of polychlorinated alkanes were obtained. This material was identical to the material isolated in the endo case in analysis and infrared spectrum. , It would arise by a scheme similar to equation 20. Examination of the infrared spectrum of the crude products indi- cated a peak at 5. 58 u. 1 It is unusual to have this peak since the stereo- chemistry is not correct for lactone formation. There is a possibility that the material giving rise to this peak is 5-exonorbornenecarbonyl chloride (from which the peroxide was prepared). There are a number of reasons for discounting this possibility: 1) the peroxide solution was washed with 5% sodium carbonate and the peroxide recrystallized prior to decomposition; 2) the intensity of the 5. 58 111 band increased as the decomposition proceeded; 3) the amount of material possessing the 5.158 (.1 peak is the same using different batches of peroxide. A second alternative is that the lactone is the same as that formed from the endo peroxide, and is due to contamination of the exo peroxide by the endo isomer. Using the extinction coefficient of the lactone obtained from 5-endonorbornenecarbonyl peroxide, the yield of lactone produced by the exo isomer was 15%. This would necessitate a minimum of 37% of the endo peroxide as a contaminant. A 10% impurity of 5-endonor- bornenecarboxylic acid in 5-exonorbornenecarboxylic acid causes the mixture to liquify at room temperature and the exo acid used to prepare the peroxide melted sharply at 43-440. Also, analysis of methyl 5-exo- norbornenecarboxylate (prepared by treating the acid with diazomethane) on a Beckman GC-2 vapor phase chromatograph indicated less than 1% Of the endo isomer. ~ A third alternative is rearrangement of the radical produced after the initial attack of trichloromethyl radical. The stereochemistry would then becorrect for lactone formation. 23 ‘ CC13 ‘ c:c13 £02 JUL 20... I. A rearrangement of this type is not without precedence. Berson (6) has shown good evidence of a free radical Wagner-Meerwein rearrangement in his work on 2, 2'-bis-azocamphane. Using the above products, a mechanism can be written: ' (25) 9 k1 c013 CO 2 + CC13 ‘- ——-> P. ‘ 9 , c-o—o-C 24 ca, cc13 P1 101 . CaO-O-C . cc1 ‘ 3 q " + co, + b c - 6613 1 o .o e 1.. .1 g \ - d - ‘C-o—O-c' + cc1, ——-1—-> c1 ’. c o o . 2 00,- ifi—s c201,, . The acid structure B could arise by a variety of methods. Homolytic . fission of the product of kg (or of the product with the chlorine and tri- chloromethyl groups reversed) followed by hydrogen abstraction, or hydrogen abstraction of the acyloxy radical from k, followed by addition of the elements of carbon tetrachloride across the double bond. . Using steady state approximations, the rate expression is as equation 18 with ki = k, «.1 kl/ka . ~ « 11. Rates of Decomposition of Diacyl Peroxides The rates of decomposition of the diacyl peroxides were followed by titrating the peroxide present at various time intervals (iodometrically), and by measuring the rate of disappearance of the 5. 63 u peroxide peak in the infrared. The disappearance of the 5.63 p. peroxide peak of 5-endonorbornenecarbonyl peroxide could not be followed due to interference by the appearance of the 5. 58 u lactone peak. 25 In the decomposition of the unsaturated peroxides, the appearance of ester was followed by measuring the rate of change of the 5. 77 11 ester peak in the infrared. .The appearance of acid was followed in the decompo» sition of the saturated peroxides by measuring the rate of change of the 5. 87 p. acid peak. The rate data and the activation parameters for the various decompositions are summarized in Tables 5, 6, 7, and 8. The errors denote mean deviations and not probable error. The measurements were carried out in the following manner. . Nitrogen was passed successively through Fieser's solution, saturated lead acetate solution, concentrated sulfuric acid, and finally potassium hydroxide pellets. A 0. 05 N solution of the peroxide in carbon tetra- chloride was purged of oxygen by passing the purified nitrogen stream through it for 10 minutes. The solution was placed in ampoules (c. a. 5 m1.) and the ampoules sealed at -700. The ampoules were placed in an oil bath controlled to i 0. 20 C. Ten minutes were allowed. to equilibrate the ampoules to bath temperature. The ampoules were removed at various time intervals, quenched in ice water, and stored at -700 until completion of the run. The tips of the ampoules were broken, the ampoules drained, and the appropriate measurements made. 2‘—Endonorbornanecarbonyl peroxide deComposed seven times slower than the corresponding exo isomer although their respective energies and entropies of activation are within experimental error. The endo isomer is also a factor two slower than cyclohexylcarbonyl peroxide and a factor four slower than cyclopentylcarbonyl peroxide (47). This seems to indicate that carbon-carbon bond stretching is important in the decomposition and steric crowding inhibits carbon-carbon stretching in 2-endonorbornanecarbonyl peroxide. This is further supported bythe production of 73. 7% carbon dioxide and 11.2% acid from the endo isomer while the exo isomer-produced 79. 3% carbon dioxide and 6 . 2% acid. 26 Table 5. Rate Constants and Activation Parameters for the Decomposition of Various Peroxides in Carbon Tetrachloride Determined by Titrimetric Techniques. Compound T/.OC k x 103 minutes.1 Ea/kcal./mole AS*/e.u. : r 44.5 $7840.16 53.9 7.24 $0.42 co 62.7 23.53:“? 24. :1: 1 3.5040.55 2 2 65.9 43143.3 1’ 44.5 1.33 40.06 53.9 2.7liO.17 24. :1 l.655:0.l4 65.9 14.2 5:0.1 44.5 3.95 «1: 0.31 53.9 7.2840371 20. 3:2. -6.2841.20 65.9 30.5 41.0 1 * 44.5 1.55:1:0.09 53.9 7.2240.35 33. :L-Z. 31.844.0 . 65.9 42.6 :1: 1.9 44.5 0.367 t 0.013 53.9 1.70:1:0.10 30. 41 15041.7 65.9 7.50 :1: 0.40 * 44.5 0.544 4 0.077 53.9 2.60:t0.08 25. i3. 0.32:1:0.09 ' 2 65.9 7.70tl.ll 44.5 2.81 £0.60 53.9 12.340.8 28. i6. 17.5i7.4 * 44.5 4.3240.35 ' 53.9 10.2:1: 1.8 24. t4. 6.4242.26 65.9 50.9 411.0 *0. 24 M Styrene present. 27 Table 6. Rate Constants and Activation Parameters for the Decomposition of Various Peroxides in Carbon Tetrachloride Determined by Infrared Techniques. :1: Compound T/OC. kx103/minutes‘l Ea/kcal./mole AS /e.u. 44.5 4.45 40.04 53.9 10.0tl.2 24.743.0 6.14:1:1.48 coz)z65.9 50.5 41.6 1 ' * 144.5 2.354004 53.9 7.9540.70 27.942.5 15.1642.66 C02 265.9 37.240.8 ' 44.5 0.297 :1: 0.027 53.9 1.8040.05 32.8-43.0 22.0-14.0 0032659 8.10 40.43 . * 44.5 0.386 40.056 53.9 2.10:1:0.08 30.244.4 14.3:1:4.1 cog—)z65.9 8.00 40.82 44.5 3.38 40.30 I 53.9 10.4-40.5 27.142.4 13.2434 co2 Z65.9 50.440.9 _ * 44.5 2.62 40.11 53.9 11.740.8 31.342.1 26.1-1:3.6 . CO; 265.9 60.243.8 :9: 0 . 2 M Styrene. present. Table 7. 28 Rate Constants and Activation Parameters for the Appearance of Ester from the Decomposition of Various Peroxides in Carbon Tetra- chloride Determined by Infrared Techniques. W 41 Compound T/OC. k x 103/minutes"l Ea/kcal./mole AS '/e.u. 44.5 5.0140.61 53.9 8.2940.05 22.342.7 -1.3040.32 62.7 28.2 41.7 C02 9265.9 45.941.6 44.5 1.36 40.03 53.9 3.7840.19 25.241.3 5.3040.54 C022 65.9 16.340.8 44.5 5.38 40.51 53.9 15.340.5 22.442.1 -0.7440.14 CO; 9265.9 49.240.1 44.5 2.30 40.08 53.9 10.541.7 29.2-44.7 19.1046.2 65. 9 41.14 0.8 a1: 0. 2 M Styrene present. 29 Table 8. Rate Constants and Activation Parameters for the Appearance Y of Acid from the DeComposition of Various Peroxides in Carbon Tetra- chloride Determined by Infrared Techniques. * Compound T/OC kx103/minutes"l Ea/kcal./mole AS l/e.u. 44.5 0.353 4 0.020 53.9 1.8040.21 30.643.6 15.543.6 CO; 2 65.9 7.60 4 0.08 * 44.5 0.520 40.028 53.9 1.4040.18 29.643.8 12.543.2 GOA} 65.9 9.80 4 0.07 1 44.5 3.65 4 0.53 ~ 53.9 10.840.2 27.544.0 15.244.4 C05); 65.9 65.94 9.1 1 * 44.5 5.1140.31 ’ 53.9 16.242.8 29.644.9 21.247.2 C0293, 65.9 79.347.2 :11 0. 2M Styrene present. 30 These facts indicate that the reaction represented in equation 26 proceeds with more difficulty than that in equation 27. (26) H A o .; go. __, ,3 H ~ (27) H In equation 26, the departing carbonyl function must move under the ring to create a trigonal carbon at position 2. This would cause severe crowding. In equation 27, the-departing carbonyl function moves toward the. 7 position which is sterically more favorable. _ PinCock (63) in his study of the t-butylperoxyesters of Z-exo-and endonorbornanecarboxylic acids found that the endo isomer was a factor four slower than the exo isomer. This rate difference and also the factor 7 found in the diacyl peroxides is too small to consider the formation of a bridged free radical. ..O .1 A C 0‘ > “J + co, (28) , The unsaturated peroxides decomposed at approximately the same rate as Z-exonorbornanecarbonyl peroxide. This also supports the concept of steric crowding in 2-endonorbornanecarbonyl peroxide. 1 Cooper (65) found that dibut-3-enoy1 Speroxide[(CHz=CH-CH2-COz-fzndecomposed fifty times faster than dibut-Z-enoyl peroxide[(_CH3-CH=CH-COZ+7)]- Cooper attributed the rate difference to the production of an allylic radical in the 31 former case. Double bonds further down the chain as in dioleoyl peroxide [(CH3(CHz)-,CH=CH(CHZ)7COZj-z] had very little effect on the rate of decomposition. Since the unsaturated peroxides decompose at very nearly the same rate, and 3-ch1oronortricyclene was not detected in the products, the double bond must not participate in the decomposition of 5-exonorbornene-a carbonyl peroxide 1 4 in. __. /'J° Styrene had very little effect on the rates of decomposition of the + CO2‘ (29) 1 saturated peroxides indicating that the ki term in equation 18 for these peroxides is very small and the observed rate constants are'very close to the rate of Spontaneous decomposition. Styrene was found to inhibit the rates of decomposition of the unsaturated peroxides but was not effective enough to completely halt the induced decomposition. EXPERIMENTAL 1. Apparatus and Reagents A. A aratus The apparatus used to determine the amount of carbon dioxide produced in the decompositions of the peroxides is illustrated in Figure 1. All infrared spectra were taken on a Perkin-Elmer model 21 record- ing infrared spectrophotometer. Unless specified otherwise, all infrared spectra were taken in carbon disulfide solution except for the region of 6. 2-7. 2 uwhich was taken in carbon tetrachloride solution. Solvent evaporations were performed using a Rinco type rotary evaporator. B. Purification of Carbon Tetrachloride Reagent grade carbon tetrachloride was purified using the method of Teller (20). A mixture of 30 g. of potassium hydroxide, 180 ml. of 95% ethanol, and 180ml. of water was added to 2. 5 1. of carbon tetrachloride. The mixture was heated at 600 for 30 minutes, the layers separated, and the process repeated. The carbon tetrachloride was washed twice with water to remove the ethanol, then washed with small portions of concen- trated sulfuric acid until the sulfuric acid layer was clear. After washing twice more with water, the carbon tetrachloride was dried over calcium chloride and distilled from phosphorus pentoxide through a 30 x 1 cm. glass helix packed column. . The distillate boiling at 760 was collected. 32 33 .mcofldfigu0uofl 000530 £03.30 MOM p00D madnmmm< 0%. mo Emuwgfl .H 0ufimmh .033 0.3.030 04,3000on .NH .093 0coup>€010fiumomm 0>303oun~ .: .033 0aonp>€mnofiuoomm 00.58 .3 .mmms 00w1>nQ .o .0o0mmoam 0>o50u o» 053:4 .w 4000044 Go$000m .v .0H5u0u0m500 00.3000 «0 “SEER—came 03:05 92303 .o .c0monfi: Bonfire oo. 03.830030» conned wcwfimucoo 0G0 0030.980 5%? 0093100 Mmdfim .m .300 0353.90 045500 00 .0pwxoupeas 5.300000% .0 .o0woufio >30 0» .300 03330 00000300280 .m .mpodomEoo H5350 0>o§0n o» £000? a: 030000 002 00000.3.mm .16. £20?wa 0>o§0u 3 4.83300 01000fm .d 2 3 0H m m N. o m 0 m .N H nousnououo. Soil/4o / \Q‘ Howoo . .\. . 1.19} 0.3 s W O, .0. D ‘8‘ ‘ O. \D ‘1 .0 u HOMO\\P film. 4 W” m & rvl "on. n . 4 1 a 50 ...z 1|. L. 3 NZ A Y1 34 C. Standardization of Sodium Thiosulfate The method of Silbert and Swern (36, 37) was used to determine the purity of the diacyl peroxides. To test the method for applicability to diacyl peroxides, the sodium thiosulfate solution was standardized using potassiumiiodate and highly purified benzoyl peroxide as primary standards. ApprOpriate amounts of the above reagents were placed in 125-ml. iodine flasks and a small piece of dry ice added to purge out the oxygen present. Acetic acid (5 ml.) containing 0. 0005% ferric chloride hexa- hydrate, and O. 5 m1. of saturated sodium iodide solution was added. The mixture was allowed to stand in the dark for 25 minutes° Water (20 ml.) was added, and the magnetically stirred mixture was titrated with ca. 0. 01 N sodium thiosulfate. Starch solution (5 ml.) was added near the end point and the end point was taken at the disappearance of the purple starch- iodine color. The results obtained from each of the primary standards agreed within experimental error (see Tables 9 and 10). II. Preparation of Diacyl Peroxides A. Preparation of Acids 1. Preparation of 5-Endonorbornenecarboxylic Acid The method of Alder (38) was employed in this preparation. Freshly distilled cyclopentadiene (34 g. , 0. 51 mole) was added to ice cold acrylic acid (40 g. , O. 55 mole) contained in a 125-ml. Erlenmeyer flask. . The re- Ws‘kewwuttevdw ' actionmixture was cooled/(in an ice bath to ensure a temperature maximum of 400. - On occasion the reaction will proceed withnenough vigor to cause the reaction mixture to boil, but usually the reaction proceeds smoothly to completion in about 25 minutes. The mixture was dissolved in 5% sodium carbonate solution and washed with ether to remove neutral residues. The aqueous solution was acidified with 6M sulfuric acid and extracted 35 Table 9. Standardization of Sodium Thiosulfate Using Potassium Iodate as a Primary Standard. ”i Sample 1 2 ‘ 3 " 4 5 Blank O Titer in 16.42 16.40 16.42 16.42 16.40 0.00 Milliliters Normality in 0.01309 0.01311 0.01309 0.01309 0.01311 Equivalents / Liter Average normality 0. 01310 i 0. 00004. Table 10. Standardization of Sodium Thiosulfate Using Benzoyl Peroxide as a Primary Standard. - Sample 1 2 3 4 5 Blank Titer in 16.70 16.68 16.65 16.65 16.67 0.00 Milliliters ' Normality in 0.01315 0.01317 0.01318 0.01318 0.01318 Equivalents / Liter Average normality 0. 01317 i 0. 00004. 36 with ether. After drying over magnesium sulfate and evaporating the ether, the residue was distilled i_1_1 wthrough an 8" Vigreux column to yield colorless acid, b.p. 100-102°/2 mm (lit. value (39), b.p. 113.54120.5°/ 5. 7 mm. ). The acid was then crystallized from pentane by cooling in dry ice to yield 50 g. (0. 362 mole, 71%) of white 5-endonorbornenecarboxylic acid, m.p. 43-44° (lit. value (40), m.p. 45-460). 2. Preparation of Methyl 5-Endonorbornenecarboxylate The method of Roberts (41) was employed to prepare this compound. A 300-ml. three-necked flask fitted with a mechanical stirrer, reflux condenser, and a dropping funnel was charged with 65 g. (0. 75 mole) of methyl acrylate (b.p. 79-800/1 Atm.), 0. 5 g. of hydroquinone, and 50 ml. of anhydrous ether. The reaction mixture was cooled in an ice bath and 45 g. (0. 68 mole) of freshly distilled cyclopentadiene was added dropwise with stirring over a 2 hr. period.’ The mixture was stirred at ice bath temperature for an hour and then an additional hour at room temperature. The ether was evaporated and theresidue fractionated i_1_i flip through a 30 x 1 cm. glass helix packed column to yield 95 g. (92%) of colorless ester, b.p. 58°/4.2 mm., n25 1.4726 (111;. values (41), b.p. 63.5°/5.2 D mm., n31.4718). 3. Preparation of 5-Exonorbornenecarboxylic Acid I The method of Roberts (41) was employed in this preparation. A mixture of 65 g. (0.45 mole) of methyl 5-endonorbornenecarboxylate, 39 g. (0. 72 mole) of sodium methoxide, and 91 g. of absolute methanol con- tained in a.500-m1. flask was refluxed on a steam bath for 48 hours. Most of the methanol was then stripped off at aspirator pressure using an ebullator to reduce bumping. Water (50 ml.) was added and the mixture refluxed for 24 hours. The methanol formed was removed by fractionation through an 8" Vigreux column at atmospheric pressure and the residual 37 solution washed with ether. The aqueous layer was acidified with 6 M sulfuric acid to Congo Red and extracted with ether. After drying over magnesium sulfate and evaporating the ether, the residue was distilled i_n wthrough an 8" Vigreux column to yield 41 g. (70%) of colorless acid, b.p. 100-1020/1.5 mm.(1it. value (41), 103.5-104°/2.2 mm.). The crude acid was purified according’to the method of Ver Nooy and Rondestvedt (42, 43). The crude acid (41 g. , 0. 297 mole) was neutralized with 10% sodium hydroxide in a 1-1. separatory funnel. . Sodium bicarbonate (7 g.) and excess iodine solution (about 900 m1. of a solution 0. 67 M in iodine'and 2. 0 M in potassium iodide) was then added. The dark oil which formed was extracted with ether. The ether extracts were combined, washed with 10% sodium thiosulfate, and dried over calcium chloride- Evaporation yielded 36 g. (40%) of crude gamma-3 lactone of 2-exoiodo-3-endohydroxy 5-endonorbornanecarboxylic acid. The aqueous layer was treated with 10% sodium thiosulfate, acidified with 6 M sulfuric acid, and extracted with ether. The ether extracts were combined, washed with 1% sodium thiosulfate, and dried over magnesium sulfate. After evaporating the ether, the residue was dis- tilled i_r_1 \La_c_upthrough an 8" Vigreux column to yield 24 g. (58. 5%) of white acid, b.p. 100-1010/1.5 mm. The 5-exonorbornenecarboxylic acid was recrystallized from pentane by cooling in dry ice, m.p. 43-44o (lit. (42), b.p. 100-101°/1.25 mm., m.p. 44.—45°). 4. Preparation of 2-Endonorbornanecarboxylic Acid 5-Endonorbornenecarboxylic acid (85 g. , 0. 616 mole) was dissolved in 250 m1. of ethyl acetate and hydrogenated over 0. 5 g. of 5% palladium on charcoal at room temperature. Initial hydrogen pressure was 50 pounds per square inch and theoretical hydrogen uptake was complete after 1 hour. The palladium on charcoal was filtered and the ethyl acetate evaporated. The residue was distilled i3 vacuo and the distillate crystallized from 38 pentane to yield 82 g. (94. 5%) of white 2-endonorbornanecarboxylic acid, b.p. 88°/O.7 mm.,. m.p. 64-660 (lit. values (38,44), 64-660, 65°). Its infrared Spectrum is shown in Figure 2. 5.. Preparation of 2-Exonorbornanecarboxylic Acid .5-Exonorbornenecarboxylic acid (40 g. , 0. 29 mole) was dissolved in 250 ml. of methanol and hydrogenated over platinum oxide at room temperature. Initial hydrogen pressure was 50 pounds per square inch and theoretical hydrogen uptake was complete after 1 hour. The platinum oxide was filtered and the methanol evaporated. The resultant oil was crystallized from pentane-ether by cooling in dry ice to yield 32 g. (79%) of white 2-exonorbornanecarboxylic acid, m.p. 56-57O (lit. values (44,45), 58-58. 50, 56-570). . Its infrared spectrum is shown in Figure 3. B. Preparation of Acid Chlorides .All acid chlorides were prepared by the method of Boehme (40). . In a typical preparation, 5-endonorbornenecarboxylic acid (10 g. ,. 0. 0725 mole) was added to 11. 9 g. (0. 10 mole) of thionyl chloride and 25 g. of chloroform contained in a 100-ml. round-bottomed flask. The flask was fitted with a reflux condenser, and after the initial reaction had subsided, the reaction mixture was refluxed for 3. 5 hours on a steam bath. The chloroform was fractionated through an 8" Vigreux column. , The pressure was lowered and 10. 3 g. (90%) 0f colorless 5-endonorbornenecarbonyl chloride was collected, b.p. 470/1. 5 mm. (lit. value (40), b.p. 84-85/ 14 mm. ). The acid chloride (c. a. 1 g.) was then treated with 5% sodium hydroxide solution. The solution was acidified to Congo Red with 6 M sulfuric acid and extracted with ether. After drying over magnesium sulfate and evaporating the ether, the residue was crystallized from 39 «1 NH .304 03>Xon0000c0snonuocopcmIN mo 533025 000.00%: .N 005wflh much 03% cw gumc0a0>0>> OH w o “v —-41 m 2_ .3 40 a; Na A: 000.4. 03>x0£0000c0auonhocoxm1N mo 8.9.3.025 6000.32; 380032 cw #3203030? w o w .\ .m 00.93am .— ‘ 41 pentane by cooling in dry ice to yield a colorless acid melting at 43-450. This acid was found to be identical to the starting acid and illustrates the absence of isomerization in the formation of the corresponding acid chloride. The properties of the acid chlorides are summarized in Table 11. . C. Preparation of Diacyl Peroxides A11 diacyl peroxides were prepared by the method of Wyman (47). In a typical preparation, 5-endonorbornenecarbonyl chloride (0. 624 g.‘, 0. 004 mole) was added to an ice-cold slurry of O. 156 g. (0. 0012 mole) of sodium peroxide and 40 ml. of anhydrous ether contained in a 300-ml. 2-necked round-bottomed flask. , The flask was fitted with a reflux condenser and a thermometer. The mixture was cooled in an ice bath and stirredmagnetically throughout the reaction time. Water (2-4 drops) was added to initiate the reaction. A rise in temperature of 1--2o is noticed upon addition of the water. A few drops of water were added intermittantly through the course of the reaction. The reaction was taken as complete when the yellow sodium peroxide color was replaced by white sodiumichloride (about 5 hours). Ice water (15 ml.) was then added and the layers separated. The ether layer was waShed successively with ice water, 5% sodium carbonate, and finally ice water again. After drying over magnesium sulfate and evaporating the ether, the residue was taken up inpentane. After drying over drierite and evaporating the pentane, the peroxide solidified. The solid was taken up in pentane and the solution cooled in dry ice. , Filtration yielded 0.48 g. (87%) of white 5-endonor- bornenecarbonyl peroxide powder, m.p. 45-470 with evolution of gas. Titration indicated 99+% purity. The properties of the peroxides are summarized in Table 12, their infrared spectra are shown in Figures 4, 5, 6, and 7. 42 Table 11. Yields and Physical Properties of Various Acid Chlorides. —‘ @ 4 hours C1 ~~E J *- Compound 4 Reaction Time Yield b.p. . Literature Value 0 84-850/14 mm. 3.5 hours 90% 47 /1.5 mm. . c-c1 (40) 6’ . o 0 70-72 /8Imm. 3.5 hours 66% 44-46 /3 mm. (40) CCl 0 840/12 mm. 4 hours 79% 82 /10 mm. c-c1 (38) / O/ ' o 0 83-84 /12 mm. 75% 83 /10 mm. (46) 43 Table 12. Yields and Physical Properties of Various Diacyl Peroxides. H— Compound Reaction Time Yield m. p. Purity Figure > 5 hours 87% 45-47° 99+% 4 co / 2 o’ P 4.5 hours 53% 49. 5.50.50 97% 5 do). 0 7 hours 60% 87 99+% 6 600}, 9 5 hours 72% 55—5o° 94% 7 co)2 44 NH H0Ufixou0na Ahconn0u0c0cuonuocopflMum mo 85.300mm 00.32%: maouowz a“ Auwa0~0>0>> ca w o .w 0udwfih — — ’\ .03Xon0m 6%Goah000c0cyonnocoxnm4m mo 85.300mm 0000.32: .m 005w?“ 0000032 3 #3030403 3 w o v 45 _ _ _ _ 46 \ .0pfixOn0nH H.»Gonu000c0cpon:ocopcmIN mo E50325 p0u0HwGH .0 0.30me 02000022 a“ #3203050? 1: NH OH w o v _ _ _ B 2 _ r1 _ _ _ _ _ + 47 0pr0.0% H>Gonp000c0nhonnocoxmum mo 85.300mm 0000.5“; 0:00 0344 .5 #3203030? OH w o . N. 0udmflh «40* H —h _ _ 48 III. Preparation of Lactones A. Preparation of the Gamma-3 Lactone of 2-Exobromo- 2-endohydroxy-5-endonorbornanecarboxylic Acid The procedure of Roberts (41) was employed in the preparation of this compound. 5-Endonorbornenecarboxylic acid (10 g. ,. 0.0725 mole) was dissolved in a solution of 30 g. of sodium bicarbonate in 450 ml. of water contained in a 500-m1.- Erlenmeyer flask. The mixture was cooled in an ice bath and stirred magnetically. Bromine (3. 8ml. , 0. 0736 mole) was added dropwise to the stirred solution. - A light yellow oil separated and) the mixture was extracted with ether. 7 The ether solution was washed with. 10% sodium thiosulfate and finally with water. . After drying over calcium chloride, the ether was evaporated and the residual oil crystallized from ethyl acetate-pentane by prolonged cooling in dry ice. The solid was _ filtered and recrystallized to yield 4 g. (25%) of the gamma-3 lactone of Z-exo- bromo-3- endohydroxyaS-rendonorbornanecarboxylic acid, colorless needles, m. p. 66-670 (lit. value (41), m.p. 67-680). - Its infrared and mm. r. spectra are shown in Figures 8 and 32. B- Preparation of the Gamma-3 Lactone of 2—Exochloro- 3-endthdroxy-5-endonorbo rnanecarboxylic Acid 5-Endonorbornenecarboxylic acid (5 g. , 0.0363 mole) was dissolved in a solution of 15 g. of sodium bicarbonate in 225ml. of water. . The magnetically stirred solution was cooled in an ice bath and chlorine gas was passedvin until the solution took on a green-yellow color. The mixture was extracted with ether, and the ether solution washed successively with water, 10% sodium thiosulfate, and finally water again. 7 After drying over calcium chloride and evaporating the ether, the residual oil was crystal- lized from pentane-ethyl acetate by cooling in dry ice. The solid was filtered and recrystallized to yield 3 g. (37%) of lactone, m.p. 75-760. 49 uaoaopa0umu>xoup>£opa0umsoaounoxMuN Mo 95004 mn0§§00 05 mo 93.3035 0000.35 OH mcouoflz a“ auma0fi0>0a$ m .304 uflcnxonu000c0cnon .0 $me _> _ 50 Its infrared and n.m. r. spectra are shown in Figures 9 and 32. ’Anal. .Calc'd for 0315190201: c, 55.65; H, 5.26; CI, 20.56. Found: C, 55.01; H, 5.14; Cl, 21.92. C.~ Preparation of the Gamma-3 Lactone of 2-Exoiodo-3-endohydroxy- 5—endonorbornanecarboxylic Acid This compound was obtained in the crude form in the preparation of 5-exonorbornenecarboxylic acid. The crude material was recrystallized twice from ethyl acetate-pentane by cooling in dry ice to yield the lactone, m.p. 58-590 (lit. value (77), 58-590). Its infrared and n.m. r. spectra are shown in Figures 10 and 32. - IV. Preparation of Esters A. Preparation of Alcohols 1. Preparation of 5-Endohydroxynorbornene 51-Endoacetoxynorbornene was prepared by the method of Alder (48). Freshly distilled cyclopentadiene (25 g. , 0. 378 mole) was added to 37. 5 g. (0.436 mole) of vinyl acetate in a combustion tube and sealed. The tube was placed in an autoclave and heated at 1900 for 12 hours. . The contents were distilled i_r_1 y_a_c_1_1£to yield 18 g. (30%) of colorless 5-endoacetoxy- norbornene,. b.p. 800/15 mm. , ing 1.4685 (lit. values (41), b.p. 720/10 mm. , n3 1.4668). 2 The 5-endoacetoxynorbornene (8 g. , 0.0526 mole) was added to 125ml. of 10% sodium hydroxide contained in a 250-ml. round-bottomed flask fitted with a condenser, and the mixture refluxed for 12 hours. . The reaction mixture was extracted with ether. ‘After drying over mag- nesium sulfate and evaporating the ether the residue was sublimed in vacuo, 1000/10 mm. The sublimed solid was crystallized from pentane by cooling 634 ofiraxonumoocmcuoo. unocopcouma>xOuU>A0©aoamnouozooxmum Ho occaomq mumggmo 04.. mo Eauommm bonanwcH .o oudmfim mcouoflz GM AuwGoHon? NH OH m o w. 51 _ _ _ 4 _ _ 52 634 oflbnonumomcmcuon. , unoccuamumu.»x0ub>£0UConmuopomoxMuN mo occuomq mumegmu 66.3 mo Efinuooaw pendumfi mcou 03% ca prsmamcrm? NH OH w o .2 356E ‘ _ _ _ w fl 53 in dry ice to yield 2. 7 g. (46. 7%) of white 5-endohydroxynorbornene, m.p. 109-11o° (lit. value (41), m.p. 109.4-110.8°). The 3, 5-dinitrobenzoate was prepared by the procedure of Applequist (49). Colorless needles from pentane-ethyl acetate, m.p. 105-1060. - Its infrared spectrum is shown in Figure 11. ~ Anal. . Calc'd for CMHIZNzOéz‘ C, 55. 26; H, 3. 94. Found: C, 55.16; H, 4.09. 2. Prejaration of 5-Exohydroxynorbornene S-Exohydroxynorbornene was prepared as described by Roberts (41) using the method of Doering and Aschner (50). 5-Endohydroxynorbornene (2 g. ,. 0. 0185 mole) was added to a solution of 0. 05 g. of fluorenone in - 5 m1. of toluene contained in a 10-ml. round-bottomed flask fitted with a reflux condenser. ‘ A small piece of sodium was added and the mixture refluxed for 17 hours.. The mixture was poured on 25 m1. of water and extracted with ether. . After drying over‘ magnesium sulfate and evaporat- ing the ether, the residue was sublimed i_r_1 ya_c_1_1_o_, 90°/1o mm. . The sub- limed solid was crystallized from pentane by cooling in dry ice to yield 1 g. (50%) of white spongy 5-exohydroxynorbornene,. m.p. 101. 5-103.5o (lit. value (41), 97.5-99.2°). The 3, 5-dinitrobenzoate (49) gave colorless needles from pentane- ethyl acetate, m.p. 104-1050. ~Its infrared spectrum is shown in Figure 12. £211.. Calc'd for CMleNzOé: C, 55. 26; H, 3. 94. Found: C, 55.25; H, 3.98. 3. Preparation of 2-Endoacetoxynorbornane Into a 250-ml. hydrogenation bottle was placed 12. 5 g. (0. 08 mole) of 5-endoacetoxynorbornene, 75 ml. of glacial acetic acid, and 0. 5 g“. of platinum oxide. The mixture was hydrogenated at room temperature with 54 «1 NH .mumouconoufigflum .m HencosuonuocopsMum mo 8530on 66.332: mcou 03% .5 239230.935 OH w o I. .2 eusmam _ - _ _ 55 I\ .oumouconouficflflum ..m Knockonuocoxmum mo 93.30on mononwcH .NH oudwfih muonofiz a“ gumcoamgm? NA NA OH m o w a _ _ fl ‘ _ _ ‘_ _ F _ H 56 50 lbs. hydrogen pressure. Hydrogen uptake was complete in 15 minutes. . The mixture was filtered and the filtrate poured on 250 m1- of water. , The aqueous solution was extracted with ether. The ether extracts were washed with water, 10% sodium carbonate, and finally, water once more. After drying over‘ magnesium sulfate and evaporating the ether, the residue was distilled i_r_1 w to yield 12 g. (96%) of acetate, b.p. 75~77o/ 13 mm., n3 1.4577 (reported (48), 81-830/12 mm., n3 1.4578). 4.- Preparation of 2-Endohydroxynorbornane 2-Endoacetoxynorbornane (6. 5 g. , 0.042 mole) was dissolved in 25 ml. of anhydrous ether. The, solution was added dropwise to a stirred slurry of 1 g. (0. 026 mole) of lithium aluminum hydride in 75 m1. of anhydrous ether contained in a 300-ml. 3-‘necked round-bottomedrflask fitted with a reflux condenser, mechanical stirrer, and a dropping funnel. After the addition was complete (about? hour), the mixture was refluxed for %- hour and stirred for an additional hour. . The excess lithium aluminum hydride was destroyed by adding small portions of ice. . The mixture was poured on water, acidified with 10% sulfuric acid, and extracted withether. The ether extracts were washed with 5% sodium carbonate, and finally water. After drying over magnesium sulfate and evaporating the ether, the residue was crystallized from pentane by cooling in dry ice to yield 3 g- (63%) of alcohol, m.p. 144-1450 (reported (48), m.p. 149-1500). 5. Preparationof Z-Exohydroxynorbornane The method of ,Bruson (51) was employed in the preparation of this compound. Norbornene (35 g. , 0.372 mole) was added to 148 g. of 25% sulfuric acid containedein a. 300-ml- 3-necked round-bottomed flask fitted with a mechanical stirrer and a reflux condenser. The mixture was stirred and refluxed for 5 hours. , The reactionmixture was cooled and extracted with ether. , The ether solution was washed with 5% sodium 57 carbonate and finally water. After drying over'magnesium sulfate and evaporating the ether, the residue was distilled to yield 9 g. (22%) of alcohol, b.p. 185-‘1900/1 atm. The alcohol was« crystallized from nitrou- methane, m.p. 123-123.5° (reported (51), m.p. 126°). 6. Preparation of 5-Endohydroxymethylnorbornene Methyl 5-endonorbornenecarboxylate (5 g. , 0.0329 mole) was dis- , solved in 25 m1. of anhydrous ether and added dropwise to a suspension of 1. 89 g. (0. 050 mole) of lithium aluminum hydride in 50 ml. of anhydrous ether contained in a 300-ml. 3-necked round-bottomed flask fitted with. a dropping funnel and a reflux condenser. The mixture was refluxed for 1 hour on a steam. bath, cooled, and the excess lithium aluminum hydride destroyed by adding small portions of ice. Hydrochloric acid (6 N) was added to dissolve the insoluble hydroxides and the mixture filtered. The filtrate was extracted with ether and the ether solution washed with 5% sodium carbonate and finally water. , After drying over magnesium sulfate and evaporating the ether, the residue was distilled i_r_1 w to yield 3. 0 g. (73.4%) of colorless 5-endohydroxymethylnorbornene, b.p. 98-1000/11 mm., rig 1.4949. The 3, 5-dinitrobenzoate (49) gave a yellow spongy solid from pentane- ethyl acetate, m.p. 80-810. - Its infrared spectrum is shown in. Figure 13. - Anal. . Calc'd for C15H14NZO6: C, 56.60; H, 4.40. Found:- C,. 56.56; H, 4.41. 7. Preparation of 5-Exohydroxymethylnorbornene 5-Exonorbornenecarboxylic acid (2 g. , 0. 0135 mole) was dissolved in 25‘ ml. of anhydrous ether and added dropwise to a susPension of l g. (0. 025 mole) of lithium aluminum hydride in 50 ml. of anhydrous ether contained in a 300-ml- 3-necked round-bottomed flask fitted with a dropping funnel and a reflux condenser. The mixture was refluxed for //1/\ f>l(ll\ <|/.lu(/|\l/\ — \ <\ 58 .mascuonuocgguofigxoupewflopcmam mo mumonconouficwfltm .m 9.3 mo 85.30QO poumumcH .mH oudmfih mcouoflz cw guwcoaoufimg a; Na OH m o 2v _ a _ q _ 4 59 1 hour on a steam bath, cooled, and the excess lithium aluminum hydride destroyed by adding small portions of ice. Hydrochloric acid (6 N) was added to dis solve. the insoluble hydroxides and the mixture filtered. The filtrate was extracted with ether and the ether solution washed with 5% sodium carbonate and finally with water. After drying over magnesium sulfate and evaporating the ether, the residue was distilled i_n vacuo to yield 1.2 g. (66.3%) of colorless 5-exohydroxymethylnorbornene, b.p. 94-95°/11mm., n3 1.4972. The 3, 5-dinitrobenzoate (49) gave colorless needles from pentane- ethyl acetate, m.p. 98-990. - Its infrared spectrum is shown in Figure 14. én_a_1_. . Calc'd for C15H14O6: C, 56.60; H, 4.40. Found: C, 56.45; H, 4.27. B. Preparation of Esters All esters were prepared in the same manner (49). - In a typical preparation, 2-endonorbornanecarbonyl chloride (1.4 g. , 0.0087 mole) was added to 1. 12 g. (0. 0094 mole) of 2-endohydroxynor- bornane in 50 ml. of carbon tetrachloride contained in a 100-ml. round- bottomed flask fitted with a reflux condenser. . Pyridine (4 ml.) was added and the mixture refluxed for 2 hours on a steam bath. The reaction rmixture was cooled and poured on 70 g. of ice. The mixture was extracted with ether and the ether washed successively with 3 N hydrochloric acid, 5% sodium carbonate, and finally water. After drying over magnesium sulfate and evaporating the ether, the residue was sublimed i_r_1 2232' 1100/11 mm. The sublimate was crystallized from pentane by cooling in dry ice to yield 1. 3 g. (64%) of ester, m.p. 109-110°. >Its infrared spectrum is shown in Figure 17. The extinction coefficient of the 5. 79 p. carbonyl band was found to be 470 l./mole-cm. £821. . Calc'd for'C15szOZ: C, 76.88; H, 9.46. Found: C, 76.84; H, 9.47. The properties of the esters are summarized in Table 13, their infrared 7 spectra are shown in Figures 15, 16, 17, and 18. 60 .osmcnonynoflfixpfiogxonp>£oxm..m mo oumoucmnouficwflum .m 23 mo Edhuoomm peyote: .va ondwfim mconoflz Ga apmnoamim? 2: NH 0H m b w _ a _ . A A 61 oawa as are ewe om m so.o so.o no ..ONH he. - e o: . . S 0:. mooH cave om w “@900 o nwoo c U I 0 mp . . . m 5: m5. MUN“. exam was mm m mwoo o mmvoo o - /U .. 01w o . . . m: mow. um.oo 5o 2» mm M omoo o mmoo o .Euumaog\muoufl, as earn a t. .m :00 2:355 3:82 62830 oufimfim mo uqmmoflwoou . £75 33% mo no m0 634 vasomEoU Goflonmuxm . muofidzflz muofifidflz moaoz m0 9332 mnoumm mdofinm> mo mmflpmmoum Adowmenflnm pad 3:3? .mH £nt 62 a: .3.353023oococuonyuosopcoum Hencocnonnocopcmtm mo 8.930on pmumpwda NH OH mcou 032 d“ ganged/mg w .2 $de _ — 63 NH .mudaxonumnvoaocuonuocoxm..m H>amanonuocoxnmtm mo 5530on @9333 OH mnon 034:“ ”3953583 w .2 Sara fi b 64 .mudH>XOQumuoamcuonuoaopcmIN Henauoeflocopcmnm Ho 9353on beams”: mcou 32 a: afimcgoufim? 2 m o \ .5 35mg _ _ _ 65 «l .mamaxonumoocmcuontosoxm..N Humzponhocoxm—um mo Ennuommm pendumcu muonoflz cw Auwnoaoufim? w .2 measure _ —) 66 V. Decomposition of Diacyl Peroxides A. Identification of the Products of Decomposition 1. Z-Endonorbornanecarbonyl Peroxide Products A 0. 05 N solution of 2-endonorbornanecarbonyl peroxide incarbon tetrachloride (250 ml.) was refluxed for 24 hours. The carbon tetrachloride was removed by distillation through a 600 x 7 mm. vacuum jacketed tantalum wire spiral column (bath temperature, 850). When no more carbon tetra- chloride came over, the column was removed and the pressure was decreased gradually to 11 mm. The distillate to 800 was collected. ~ Analysis on a Beckman GC-2 vapor phase chromatograph (retention time at 1000 = 9 min.) indicated one alkyl halide. Its infrared spectrum .is shown in Figure 19. This compound has been identified as 2-exochloroa norbornane by virtue of its infrared spectrum (see Figure 20). . The pot residue was taken up in ether and extracted with 10% sodium hydroxide. The ether layer was dried over magnesium sulfate-norite. After evaporating the ether, the residue was sublimed, 1000/10 mm. The sublimate was crystallized from pentane by cooling in dry ice to yield a material (0. 12 g.) melting at 106-1080. ~ Its infrared spectrum is shown in Figure 21. This compound is identical to the previously prepared 2-endonorbornyl 2-endonorbornanecarboxylate. The basic solution was acidified with 10% sulfuric acid and extracted with ether. After drying over magnesium sulfate-norite and evaporating the ether, the residue was crystallized from pentane to yield an acidmelté ing at 60-620. - Its infrared spectrum is shown in Figure 22. . This com- pound was found to be identical to 2-endonorbornanecarboxylic acid. The above ester and acid were identified on the basis of melting points, mixed melting points, and infrared spectra. 67 .0308qu T300330 nocmcuonnosopdmum mo coflfimomEoooQ 9.3 Eoumpmcflmst 302mm T934 may mo Edhuoomm woumuwdm .oH “:3me much 03% a: aumcoamuwm? NH OH w c 2v _ _ a _ _ 3:633 .58 68 L l l l 1 l 3 5 7 9 11 13 15 Wavelength in Mic rons Z-Endochloronorbornane l 1 1 1 1 L 3 5 7 9 11 13 15 Wavelength in Microns 2-Exochloronorbornane Figure 20. Infrared Spectra of 2-Exochloronorbornane and Z-Endochloronorbornane. . .mpflnononm H..»conudoo:msuonuocopcm..N mo Comfiwomfiooofl or? 808m poaflmdbo Museum 93 mo EDppoomm pendufifi .HN magmas.” mcouoflz cw ngcofiozm? 2: NH OH w o «4 69 _ _ _ D _ _ _ 70 «A .6308qu Hensonudo umcmsponuocopcmth mo sowfimomgooom ofi 80am 905.330 30¢. 93 m0 833on consume; mconuflz cw flumaoaoufim? NH OH w o w. ..8 6.33m _ w d _ ‘ 71 2 . Z-ExQnorbornanec arbonyl Peroxide Products The procedure in this case is identical to the 2-endonorbornane- carbonyl peroxide case. The alkyl halide was identical to the above case (see Figures 20, 23). .The ester (0.15 g.)was found to melt at 118-1190 (see Figure 24) and identical to Z-exonorbOrnyl 2-exonorbornanecarboxylate. The acid was found to melt at 50-520 (see Figure 25) and identical to 2-exonorbornanecarboxylic acid. . Identifications were made on the basis of melting points, mixed melting points, and infrared spectra. 3. 5-Endonorbornenecarbonyl Peroxide Products 5-Endonorbornenecarbonyl peroxide (5 g.) was dissolved in 25 ml. of carbon tetrachloride and the solution. refluxed for 24 hours. . The carbon tetrachloride was removed by distillation through a 600 x 7 mm. vacuum jacketed tantalum wire spiral column (bath temperature, 850). When no more carbon tetrachloride came over, the column was removed and the pressure dropped slowly to 15 mm. The distillate collected in this manner was concentrated by the above procedure to remove‘most of the carbon tetrachloride and then analyzed on a BecknxanMegachrom vapor phase chromatograph at 1500. The first fraction was found to be carbon. tetra-é chloride, the second fraction had a retention time of 20 minutes at 1050.' Its infrared spectrumis shown in Figure 26. This compound has been identified at 5-exochloronorbornene by virtue of its infrared spectrum (see Figure 27). . Infrared inspection of the residue from the first distillation indicated the presence of ester and possibly lactone (5. 77 u, 5. 58 (1). This material was taken up in a small amount of ether and added to 10 ml. of 5% sodium carbonate solution containing 1 pellet of sodium hydroxide. The mixture was stirred at room temperature for 16 hours and then extracted with ether. The ether solution was dried over drierite and the ether evaporated. . The infrared spectrum of the material indicated the presence of ester (5. 77 u). 72 . ogxouonm 12.53me twangaonuocoxmtm Ho Gowfimomaooofl 9.3 Sou“ watchman—O 66.3mm 17:4 053 Ho 8550QO bondage; 0H macho“: CH flawcoaourm? w .mm madman 3538 .88 _ f’\/\ 2 73 «l 6.303qu gconnmo nocmcuoauosoxMuN mo comfimomvfiooon 93 503 poEmBO poumm 93 mo $3.30on pendumza .vm oudmfim mcouoflz d“ Aumcmaoufim? NH OH w o w _ _ d _ _ 74 a; .3038qu Htwconhmo uoamcuoohocoxmum mo sowuwmomgoomfl «.5. 59¢ «5:330 Box» 23 mo 8.930QO pounds; mcouoflz cw “imaged/mg Mg Ca w o w. .3 853m ——-I) CI‘ 4 a _ 75 sococp0£HOGOUGMIm mo comfimomgoooQ 9.3 So: pmfimfiO opflmm H.534 on» mo 85.30on ponmumcH .om oudmwh NA 98.83% a: 5.983.483 w .oExOHonm Hewnoaqu l ( cerium :oov _ a 76 1 L 1 l 1 3 5 7 9 11 13 15 Wavelength in Microns 5-Endochloronorbornene 5 7 9 ll 13 15 Wavelength in Mic rons 5-Exochloronorbornene 1 1 l 1 l l I 3 5 7 9 11 13 15 Wavelength in Mic rons 3-Chloronortricyclene Figure 27. Infrared Spectra of 5-Exochloronorbornene, 5-Endo- - chloronorbornene, and 3-Chloronortricyclene. 77 The ester-containing material was treated with.20% potassium hydroxide in aqueous dioxane at reflux for one hour. . After acidification with 6 N hydrochloric acid, the solution was extracted with ether. - After drying over‘ magnesium sulfate and extracting the ether, the infrared spectrum of the residue indicated the absence of alcohol (2. 7=3. 0 p.) and acid (5. 9 p). The presence of the ester peak (5.77 p.) indicated that no hydrolysis had occurred. . A portion of the ester-containing material was distilled. .Material boiling at 92-93o/0.07 mm. was collected. - Infrared analysis (Figure 28) indicated the absence of carbonyl. _ _ Anal. , Found: C, 35.33; H, 3.68; Cl, 60.59. The remaining portion of the ester-containing material (ca. 3 g.) was dissolved in 25 m1- of anhydrous ether and added dropwise to a suspension of 1. 5 g. of lithium aluminum hydride in 50 ml. of anhydrous ether. The mixture was refluxed for 1 hour, then cooled. The excess lithium aluminum hydride was destroyed by adding small portions of ice, The insoluble basic materials were dissolved by adding 6 N hydrochloric acid. The mixture was filtered and the filtrate extracted with ether. The ether solution was washed with 5% sodium carbonate and finally with water. , After drying over magnesium sulfate and evaporating the ether, the residue was distilled at 13 mm. and three fractions were collected: 30-800 (a), 30-110° (2), 110-150° (3). The three fractions and the pot residue were treated for the preparation of 3, 5-dinitrobenzoate deriva- tives (49). . Fraction 2. gave the 3, 5-dinitrobenzoate of S-endohydroxymethyl- norbornene, identified by its melting point,. mixed melting point, and infraredspectrum (Figure 29). t The pot residue gave the 3, 5-dinitro- benzoate of S-endohydroxynorbornene, identified by its melting point, . mixed melting point, and infrared spectrum (Figure 30). . Fraction 1 gave the 3, 5-dinitrobenzoate of ethanol. Fraction- 3 was not identified. 78 .6363qu H>coaumUoGocuonuocochum mo cofimmomaooofl of Son“ pocflmuno canvzmonodflorom 23 mo 8.930on penaanH mconoflz a“ gumcofimxfim? NH ofi w 0 .mm enema . _ _ u 79 .Aoumzfimwflv opflnoumna Hensonhdomcocnonpocopamtm mo coflfimomEoooQ on» 50.3 pocflwpnO Hosoodee ego mo oumoncmnonficwflnm .m mg..— wo guuommm pendant: 98.822 5 prcoflgm? a; Na OH w o ‘ .om Semen _ _ a a _ 80 .Aosgmom pomv opwanom anonmeoaocnonuocopGMum mo somfimomgooofl 65. EoanoGQEO 33.0014. 053 m0 mumonconoufififltm .m 05 «0 85.30QO @9335 .om 0.23th 98.322 g fiwn36>m3 3 NH 2 w o w _ _ _ _ q d _ _ _ e r 1 LP? 81 Number 2,. 3,5-dinitrobenzoate, m.p. 80e8lo. ‘Anal: Calc'd for C15H14NZO6: C, 56.60; H, 4.40. Found: C, 56.70; H, 4.51. pot residue, 3, 5~dinitrobenzoate, m.p. 105-1060. Anal: Calc'd for C14H12NZO6: C, 55.26; H, 3.94 Found: C, 55.34; H, 4.05. ,Number 1, 3,5-dinitrobenzoate, m.p. 87-880. , Anal: Calc'd for C‘9H8NZO6: C, 45.04; H, 3.36. Found: C, 45.48; H, 3.41. ,Number 3, 3,5-dinitrobenzoate, m.p. 183-1840. When the pot residue from the original distillation of the decomposed peroxide solution was transferred to a molecular still (Hickman type), and the distillation continued at 1150/0. 07 mm. , the material lacking in carbonyl was distilled.» A white substance then sublimed on the sides of the still. . The white solid was dissolved in ether, the ether evaporated, and the residue sublimed, 1200/11 mm. Crystallization of the sublimed solid from ether produced a material melting at 100-1010. , This material possessed a carbonyl band (Figure 31) in the infrared at 5. 58 u. The n.m. r. Spectrum of this material is shown in Figure 32. This material has tentatively been given the structure: c13c ‘ a + :o _A_.n_a_1_: Calc'd: C, 42.35; H, 3.55; C1, 41.65. Found: C, 42.60; H, 3.70;,C1, 41.71. _ - An attempt was made to hydrolyze the chlorines. The material (32 mg.) was dissolved in. 25 ml. of 80% aqueous acetone. Silver nitrate (O. 52 g.) was added and the mixture refluxed for 2 hours“ The solution was extracted with, ether. , After drying over magnesium sulfate and evaporating the ether, the residue afforded 30 mg. of starting material. A method used by 82 .opflnononm 3:03.80 innocuonuocopcmtm mo coflfimomgoomfl ofi 89G pmcfinfio ongomd 23 mo Eauuoomm papaya; .Hm ondmfim macs 032 GM aumcofioam; 3 Na OH w o v —} _ A. _ _ _ 83 ‘7’ Value (Tetramethylsilane = 10) Figure 32. , N.m. r. Spectra of Lactones. 84 Freidlina (54) was then employed. . The material (0. 2226 g.) was warmed on a steam bath with 0. 31 m1. of fuming nitric acid for 4 hours. . Fuming nitric acid (0. 1 ml.) was added at 1 hour intervals. Water was then added and the mixture extracted with ether. . Drying over'magnesium sulfate and evaporating the ether afforded 0. 2215 g. of starting material. The material (0. 1 g.) was then added to 4 m1. of 80% sulfuric acid and warmed on a steam bath for 6 hours. Water was added and the mixture extracted with ether. . The ether was washed once with water and after drying over mag- nesium sulfate, the ether was evaporated. The residue was placed in a sublimator and heated to 760/11 mm. After 3 hours, the temperature was raised to 920 over a period of 1 hour, and then the sublimate collected at 920 was crystallized from pentane-ether. An acidic material (0. 04 g.) containing no chlorine was isolated, m.p. 119-1200. s This material has tentatively been given the structure: (B) :0 Anal: Calc'd for: C, 59.35; H, 5.53. Found: C, 60.68; H, 6.15. . Its infrared spectrum .is shown in Figure 33. 4 . 5 - Exonorbornenec arbonyl Peroxide ’ Products S-Exonorbornenecarbonyl peroxide (5 g.) was dissolved in 25 ml. of carbon tetrachloride and the solution refluxed for 24 hours. , The carbon tetrachloride was then removed by distillation through a 600 x 7 mm. vacuum jacketed tantalum wire spiral column (bath temperature, 850). When no more carbon tetrachloride came over, the column was removed and the pressure dropped slowly to 15‘ mm. The distillate collected in 85 “L .mpflnouona H>GOQHM00C0Guonnocopcmtm mo cofifimomaooofl 05 500m @001390 occuomd 03... mo mflmzoupnwm “.50 80.3 #003030 304 0:» mo 5.900025 “00.30%: mcou0fl>~ (cw numcoaonrm? NH OH w o v .2 shaman _ a _ d _ $538 anomov 86 this manner, was concentrated by the above procedure and the residue passed through a Me gachrom vapor phase chromatograph at 1500. . The alkyl halide fraction was identical with the one isolated from the endo peroxide in retention time and in its infrared spectrum (Figure 34). The infrared spectrum of the pot :residue indicated the presence of ester (5. 77 u) and also a peak at 5. 58 it. This material was treated, as in the endo case, with sodium carbonate and sodium hydroxide. The ether extracts were dried over magnesium sulfate and distilled. Distillate collected to 1600/1 mm. (Figure 35) was found to be identical to the material boiling at 92-930/0. 07 mm. in the endo case (i. e. polychloroalkanes). Material boiling at 106-2200/1 mm. was also collected. Infrared analysis of this material indicated ester (5. 77 p.) present. The ester-containing material (1 g.) was dissolved in 25 ml. of anhydrous ether and added dropwise to a suspension of lithium aluminum hydride in 50 m1. of anhydrous ether. The mixture was refluxed for one hour and cooled. The excess lithium aluminum hydride was destroyed by adding small portions of ice. The insoluble basic materials were dissolved by adding 6 N hydrochloric acid and the mixture filtered. The filtrate was extracted with ether and the ether-washed with 5% sodium carbonate and finally water. After drying over magnesium sulfate and evaporating the ether, the residue was dis- tilled. Distillate was collected up to 1200/11 mm. The distillate and the pot residue were each treated to prepare the respective 3, 5-dinitro- benzoates (49). . The distillate was identified as 5-exohydroxymethylnorbornene through. its 3, 5-dinitrobenzoate and the pot residue was similarly identified as 5-exohydroxynorbornene. Identification was based on melting points, mixed melting points, and infrared spectra (see Figures 36 and 37). Distillate, 3,5-dinitrobenzoate, m.p. 96-980. 30.23}: .Calc'd for C14H10NZO6: C, 56. 60; H, 4.40. Found: C, 56.24; H, 4.47. 87 tococuonuocoxmtm mo aoflwmomgooofl 0”: So: «0003.50 031mm 192.4 0:0 mo 85:00am ponmuwca a; Na macho“: Ga flumcofionfim? w .mgxonona 3000280 :3 enema fi fi autism Joov fl 88 .mpfixonom H>co3nmo tmcocnoniosoxmtm Ho comuwmomfiooofl 033 80.3 005.330 ocmxfiwouofiohom 03“— mo Eduuoomm 300.30de mcouoflz G3 33ma030>m3 #3 N3 o3 w o .m... enema —IL —} _ _ 89 .Aoumflflwfiflv opflxoumm 3%:o3amomcmcuo3honoxmtm mo couafimomaooofl 03p Eomm 3093.330 3030034 0.33 30 oumouco300ficfifltm .m 03..— mo Eshuoomm poumumcH 983032 a: 30mc030>m>> 31 N3 03 w o 1‘ .om oudmfih _ d fl 4 _ 90‘ .Amdpfimom wont opwxouom 3>Go3umomconno3nocoxmtm mo coflmmomgooofl 03..— 503 pocfimahro 3030034 .033 m0 oumonc03ouficflfltm .m 03» mo Eduuoomm pmumumcH . .hm oudmwh mcoaoflz a: 33mG030>m>> «3 N3 03 w o «n d _ _ _ _ _ 91 ' Pot residue, 3, 5-dinitrobenzoate, m. p. 104-1050. - Anal. . Calc'd for C14H12N206: C, 55.26; H, 3.94. . Found: C, 55.28; H, 3.99. a The water layer from the sodium carbonate-sodium hydroxide treat- ment was acidified and extracted with ether. iAfter drying over‘ magnesium sulfate and evaporating the ether, the residue was crystallized fromrether to yield an acid, m.p. 176-177. 50 (d 170) which was tentatively giventhe structures C. . C13 C ‘ Cl ‘ ' and/or ((3) C1 ‘. COZH c13c " COZH Anal: Calc'd for C9H1002Cl4: C, 37.00; H, 3.45; CI, 48.60. Found: C, 38.21; H, 3.85; CI, 49.34. , Its infrared spectrum is shown in Figure 38. The material containing the 5. 500 pt lactone bond could not be isolated. - A distillation scheme as used for the endo peroxide was unsuccessful. - As described above, the alkaline extracts gave C as the only crystalline material. , B. Determination of the Products of Decomposition 1. Determination of Carbon Dioxide Using the apparatus illustrated inFigure l, 50 ml. of a 0.05 N solution of peroxide in carbon tetrachloride was introduced into the reaction vessel and the system swept with nitrogen for 1 hour. The absorbtion tube (2:1 ascarite-anhydrone) was then weighed and put back in the system. The solution was then refluxed for a minimum of 12 half-lives under a continuous nitrogen sweep. . The heating was discontinuedand the dry ice .mpflnouona 3:03pmoococuo3uocoxmtm mo comfimomaooofl 033 E0: 3093.330 @304 033 mo Efiuuommm poumuwdH .mm 0.33me 900.0032 G3 33mG030>m>> , N3 03 w o v 92 _ _ _ _ w 3538 emov 93 traps allowed to stand at room temperature for 30 minutes; the absorbtion tube was then reweighed. . The mass gained was taken as carbondioxide liberated by the peroxide. The results are tabulated in Tables 1, 2, 3, and 4. 2.. Determination of Carbonyl Containing Products The carbonyl-containing materials that resulted from the decomposition of the various peroxides were determined by infrared spectroscopy. The results are tabulated in Tables 1, 2, 3, and 4. The amount of lactone in the case of 5-exonorbornenecarbonyl peroxide has been estimated using the extinction coefficient of the gamma-3 lactone of 2~exotrichloromethy1- 3 - endohydroxy- 5 - endonorbornenec arboxylic acid . C. Kinetics of Decomposition The rates of thermal decomposition of the diacyl peroxides were followed by: (1) Iodometric titration of samples of the peroxide solution. (2) Measuring the rate of disappearance of the 5.65 [.1 peroxide band via a Perkin. Elmer model 21 recording infrared spectrophotometer. (3) Measuring the rate of appearance of various products in the infrared region. The peroxide solutions were contained in 5 ml. Kimble Neutraglass ampoules. The ampoules were cleaned by immersingthem in hot concen- trated. sulfuric acid for 12 hours. . They were rinsed four times with water, given a final rinse with acetone and dried at 120° for 12 hours. The nitrogen used to remove oxygen from the peroxide solution was purified in the same manner as in the carbon dioxide determinations. . The following experimental procedure was used in a typical kinetic determination. About 50 ml. of carbon tetrachloride solution, .0. 05 N in peroxide, was purged of oxygen at room temperature by bubbling purified 94 nitrogen. into the solution for 15 minutes. The solution was then intro- duced as ea. 5 ml. samples into ten ampoules with a hypodermic syringe. . The ampoules were sealed at —700 with an air-gas torch. . They were placed in a metal cage and the cage immersed in an electrically heated mineral oil bath controlled to 0. 10 by means of a Fisher-Serfass Electronic Relay. A period of two minutes was allowed for the samples to reach the tempera- ture of the bath; zero time was thus assumed to be two minutes after the ampoules were immersed in the bath. At various time intervals an. ampoule was removed from the bath and immediately quenched in ice water. The ampoules were marked and stored at -700 until completion of the run. , The p tips of the ampoules were broken, the ampoules drained, and 4-ml. aliquots were pipetted into 125—ml. iodine flasks. The samples were titrated as described earlier (page 34). Using the sampling procedure described above, the rate of disappear- ance of the 5. 65 p. peroxide band and the rate of appearance of either the 5. 77 p. ester band or the 5. 87 u acid band were measured. The data were plotted as a concentration versus time curve and the method of Guggenheim (52) employed to calculate the first order rate constants. The energy of activation was calculated by aplot of log k versus the reciprocal of the absolute temperature employing the methOd of least squares (to determine the slope. The entropy of activation was calculated employing the method of Foster (53). Sample calculations for the calculation of rate constants, energies of activation and entropies of activation are given in the Appendix. 95 VI.. Mis cellaneous Experiments A. Preparation of Galvinoxyl Methylene-bis-para(2, 5-di-t-butyl)phenol (supplied by Ethyl Corporation) was recrystallized from hexane twice. The phenol (2 g... 0.00471 mole) was dissolved in 50 ml. of anhydrous ether contained in a 125-ml.~' Erlenmeyer flask and 5. 07 g. (0,0212 mole) of lead dioxide was added. “The mixture was stirred magnetically at room temperature for 4 hours, then filtered through a sintered glass funnel. - After evaporating the ether, the residue was dissolved in pentane and crystallized by cooling in dry ice. The dark blue galvinoxyl (1.1 g. , 55%) was recrystallized once and dried in a vacuum desiccator for 2 hours. B. Determination of the Purity of the Norbornenyl Acids 5-Endonorbornenecarboxylic acid and 5-exonorbornenecarboxylic acid were each treated in the following manner. The acid (1 g.) was treated with excess diazomethane in anhydrous ether. The ether was evaporated and the ester distilled, b.p. (exo) 84-850 /15 mm. , b.p. (endo) 85-860/15 mm. The esters were analyzed on a I I Beckmanemodel GC-Z gas chromatograph at 1350 (20% silicone column). Methyl 5-endonorbornenecarboxylate was found to contain less than 7% of the exo isomer and methyl 5-exonorbornenecarboxylate was found to con- tainless than 1% of the endo isomer. C. Attempted Preparation of the Gamma-3' Lactone of 2-Exotrichloro- methyl—3 - endohydroxy- 5 - endonorbornanecarboxylic Acid 5e‘Endonorbornenecarboxylic acid (5 g. , 0.036 mole) was dissolved in 50 ml. of bromotrichloromethane. The mixture was irradiated with a mercury vapor lamp for 5 hours. . The solvent was distilled and the residue was found to be polymeric. 96 Methyl 5-endonorbornenecarboxylate (5 g. , 0.033 mole) was dis— solved in 50 m1. of bromotrichloromethane and a small amount of benzoyl peroxide was added (ca. 0. 1 g.). The mixture was refluxed for 5 hours and then the solvent was distilled. The residue was refluxed with 50 m1. of 5% sodium hydroxide and then acidified. The water solution was extracted with ether. After drying over magnesium sulfate and. evaporate ing the ether, the residue was found to be polymeric. D.- Attempted Preparation of Syn- and Anti-7-Norbornenecarboxylic Acids 1. Preparation of 7-Syn-bromonorbornene 7-Syn-bromonorbornene was prepared by a method employed by Kwart (55). . Norbornene (60 g. , O. 63 mole) and 51 g. (0.63 mole) of pyridine were dissolved in 300 ml. of purified carbon tetrachloride (20). The solution was introduced into a 1-1. 3-necked roundabottomed flask equipped with a mechanical stirrer, reflux condenser, and a dropping funnel. The solution was cooled in an ice-salt bath to maintain a reaction temperature of less than 00 throughout the reaction. Bromine (102 g. , 0. 63 mole) was added with stirring over a period of 2 hours. The pyridinium bromide was filtered and the filtrate was washed with 200 ml. of 6 N hydro- chloric acid. After drying over calcium chloride and evaporating the carbon tetrachloride, the residue was fractionated in 135329 to yield 74 g. (46%) of 2, V7-dibromonorbornane, b.p. 100-1050/1. 25 mm. , n3 1. 5705 (reported (55), b.p. 70-740/0.25-0. 30 mm. , 113 1.5710). 2, 7-Dibromonorbornane (24 g. , 0. 095 mole) was added to a solution of 3. 9 g. (0. 10 g. -atom) of potassium in 120 m1. of t-butyl alcohol. The mixture was refluxed for 12 hours and poured on an equal volume of water, then extracted with ether. After drying over magnesium sulfate and evaporating the ether, the residue was distilled i_r_1 wqto yield 10. 7 g. (58%) of syn-7-bromonorbornene, b.p. 67-700/13 mm. , n?‘0 l. 5055 D (reported (55), b.p. 68-700/13 mm. , n3 1.5058). 97 2. Preparation of Syn- and Anti-7-«carbomethoxynorbornene A solution of 50 g. (0. 29 mole) of syn-7-bromonorbornene in 300 ml. of dry ether was added to 22 g. (0. 95 g. -atom) of magnesium under 200 m1. of ether. The addition was carried out over 32 hours under a nitrogen atmosphere at room temperature. Carbon dioxide was bubbled into the solution for 1 hour and then small portions of dry ice were added. _ The mixture was poured onto saturated ammonium chloride solution and acidified with cold, concentrated hydrochloric acid. The aqueous layer was extracted with ether and the ether solution was extracted with110% potassium hydroxide solution. The basic extracts were acidified with dilute sulfuric acid and extracted with ether. After drying over magnesium sulfate and evaporating the ether, 15. 8 g. of crude acid was obtained. The crude acid was treated with ethereal diazomethane, the ether evaporated, and the residue distilled i_r_1 wto yield 13.8 g. of ester, b.p. 960/27 mm. The ester was analyzed with a Perkin-Elmer model 154 vapor phase chromatograph using a 6', 30% silicone cOlumn at 1520. Four components were observed. Analysis on 12’, 20% silicone columns at 1500 (Megachrom) showed only two peaks. . Failure of the Megachrom at this point made the separation of the four components impractical. The overall yield of crude product is also too low toimake the synthesis practical. 3. Preparation of Acetyl Bromide Acetyl bromide was prepared employing the method of Bruton and Degering (56). Phosphorus tribromide (500 g. , 1. 845 moles) was intro- duced into a 1-1, 3-necked round-bottomed flask fitted with a mechanical stirrer, a reflux condenser, and a dropping funnel. . The flask was immersed in an ice bath and (341 g. (6. 92 moles) of glacial acetic acid was added over a 3 hour period. The ice bath was removed and the mixture stirred for an additional half-hour. The mixture was then distilled through a 8" Vigreux column. , All of the material boiling below 760 was collected and 98 redistilled to yield 320 g. (38%) of colorless, fuming, acetyl bromide, b.p. 74°750/1 atm (lit. value (56),, b.p. 76°). 4. Acylation of Norbornene Anhydrous aluminum chloride (149. 8 g. , 1.122 moles) and 320 m1. methylene chloride was introduced into a 1-1. 3-necked round-bottomed flask fitted with a mechanical stirrer, a reflux condenser, and a dropping funnel. The flask was immersed in an ice bath and 138 g. (1. 122. moles) of acetyl bromide in 160 ml. of methylene Chloride was added dropwise over 2 hours. The mixture became homogeneous and light orange in color. The solution was filtered through a coarse sintered glass funnel into a similarly equipped 2-1. 3-necked round-bottomed flask. The flask was cooled in an ice bath and 94. 2 g. (1 mole) of norbornene in 215 m1. of methylene chloride was added over 3. 5 hours. The solution became red in color at this point. The mixture was stirred for an additional hour at room temperature, then poured on a mixture of 125 ml. of concen— trated hydrochloric acid and 300 g. of ice. The aqueous layer was extracted with methylene chloride. The extracts were combined and washed successively with water, 5% sodium carbonate, and finally once 'more with water. After drying over‘magnesiumsulfate and evaporating the methylene chloride, the residue was distilled i_n wto yield 120 g. of a colorless liquid, b.p. 103-1050/0.8 mm. The product was found to contain bromine and had a carbonyl band in the infrared at 5. 87 u. The material decomposes on standing at room temperature. Its infrared spectrum is shown in. Figure 39. 5. Haloform-Oxidation of Agylation Product Bromine (120 g. ,, 0. 75 mole) was added slowly to an ice cold solution of 100 g. (2. 5 moles) of sodium hydroxide in 625 m1. of water. The resultant solution was added dropwise with stirring to 50 g. of the 99 00350.33 3.3004 3033 0000003002 30 003.0134 0.3... 8003 009.0330 00030030500m 033 no 8003095 0000.33: . .om 0053b 0000 032 03 3.300020%. N3 03 w o v ._ _ w 4 _ 3.838 amov 100 bromoketone obtained from the acylation of norbornene contained in 2-1. 3-necked round-bottomed flask. . The flask was fitted with a mechanical stirrer, a reflux condenser and a dropping funnel. A temperature of less than 100 was maintained over the addition period which was approximately 1 hour. After stirring for 2 hours more, the solution (now yellow) was heated at 450 until the yellow color disappeared. . The solution was cooled and washed with ether. The water layer was treated with sodium bisulfite, acidified to Congo red with concentrated hydrochloric acid, and extracted with ether. The ether layers were combined, dried over magnesium sulfate, and the ether evaporated. The acidic residue (22 g.) contained bromine and had a carbonyl band in the infrared at 5. 89 u. . Characteristic acid spread was noted in the carbon-hydrogen region of 3-4 p in the infra- red (Figure 40). 6. Dehydrohalogenation of Oxidation Product Potassium metal (54 g. , 0. 137 mole) was added to 120 ml. of t-butyl alcohol contained in a 250-ml. round-bottomed flask fitted with a reflux condenser. . The mixture was refluxed until all the potassium metal dis- solved. A solution of the bromoacid (15 g.) in 20 m1. of t-butyl alcohol was added. The mixture became tan in color and material precipitated. The mixture was refluxed for 15 hours, cooled, and poured on an equal volume of ice and 5% sodium carbonate. . The solution was washed with ether and treated with norite, after which it was acidified with dilute sulfuric acid and extracted with ether. After drying over magnesium sul- fate and evaporating the ether, the residue was taken up in ethyl acetate and treated with norite. The ethyl acetate was evaporated to yield 8 g. of a crude acid containing no bromine. Esterification with diazomethane and distillation in w produced an ester found to boil at 120-1300/15 mm. - Syn- and anti-7-carbomethoxynorbonene,has been reported to boil at 935-960/30 mm. (57). 101 .opwgonm 3304 £9.29 oGoGuonnoZ mo coflmfnoaoa 95 50.3 “55.8.30 occuoxoaoym 23 mo Goflumgxo 8.8.333 93 E03 pofidBO pwomoccoum 9.3 Ho 5530on Honda”: , .ow ondmfm mcouodz d“ £pwco~m>m>> a; Na OH m . o w _ _ _ _ A 3538 $8 102 7. Preparation of n-Butyl Lithium The method of Gilman (58) was employed in this preparation. Anhydrous ether (600 ml.) was introduced into a 2-1. 3unecked roundu bottomed flask fitted with a mechanical stirrer, a reflux condenser, a thermometer, and a dropping funnel. The system was flushed with dry nitrogen and 22.3 g. (3.23 moles) of diced lithium metal was added. About 1 ml. of a solution of 205. 5 g. (l. 5 moles) of n-butyl bromide in 200 m1. of anhydrous ether was added to initiate the reaction. The mix- ture was cooled to -100 in dry iceuacetone and the remainder of the n-butyl bromide solution added dropwise with stirring over a 2 hour period. The blue-grey solution was stirred an additional 2 hours at am0 and then filtered. 8. Preparation of Cyclopentadienyl Lithium The method of Kursanov (59) was employed in this preparation. Freshly distilled cyclopentadiene (79. 2 g. , 1. 2 moles) was introduced into a 2-1. 3-necked round-bottomed flask fitted with a mechanical stirrer, a reflux condenser, a thermometer, and a dropping funnel. n-Butyl lithium solution (prepared from 1. 5 moles of n-butyl bromide) was added dropwise with stirring. The temperature of the mixture was maintained at 0--5o over the addition time (about 2 hours). . The solution was stirred for an additional 2 hours at room temperature. 9. Preparation of the Exocyclic Enol Acetate of Acetylcyclopentadiene Cyclopentadienyl lithium was acylated employing the method of Riemschneider (60). The above solution of cyclopentadienyl lithium was cooled to 0-50 and 131 g. (1. 97 moles) of acetyl chloride was added with stirring. After the addition was complete (about 3 hours), the mixture was stirred an additional hour at room temperature. The solution was then poured on ice-10% sodium hydroxide and extracted with ether. After drying 103 over magnesium sulfate and evaporating the ether, the residue was fractionated 1_n vacuo through a 6" Vigreux column to yield 18 g. of exo» cyclic enol acetate of acetylcyclopentadiene, b.p. 105-1150/15 mm. (lit. value (60), b.p. 105115°/15 mm.). 10. Reaction of the Exocyclic Enol Acetate of Acetylcyclo- pentadiene with EthLlene The enol acetate (18 g. , O. 12 moles) was introduced into a steel bomb and charged with ethylene to a pressure of 500 pounds per square inch. The temperature was raised to 1700 and the mixture agitated by means of an internal solenoid controlled plunger for 24 hours. The pressure rose slowly to 800 pounds per square inch and then dropped slowly to 700 pounds per square inch. The bomb was allowed to cool and a black gritty solid was removed via ether rinse. After evaporation of the ether, the residue was stirred magnetically with dilute sulfuric acid for 12 hours. The reu sultant solution was extracted with ether. - After drying over magnesium sulfate and evaporating the ether, the remaining dark oil was dissolved in ethyl acetate and treated with norite. Evaporation of the ethyl acetate afforded 5 g. of an oily residue. The residue (5 g.) was dissolved in 200 ml. of dioxane and 50 ml. of 10% sodium hydroxide was added. Iodine solution (a solution 0. 67 M in iodine and 2M in potassium-iodide) was added until the iodine color perm sisted. The precipitated iodoform was filtered and the solution washed with ether. The water layer was neutralized with dilute sulfuric acid and extracted with ether. z Eb C O z 690): mgo>z Relative rate 1 8 10 . 8 11. 2 Ea(kcal. / ~ mole) 32.8 27.1 24.0 24.7 4. The production of lactone from 5-endonorbornenecarbonyl peroxide represents a novel induced decomposition involving an intra- molecular attack of a radical on a carboxyl group. 7 The production of lactone from 5-exonorbornenecarbonyl peroxide is evidence for the occurrence of a free radical Wagner-Meerwein rearrangement. 5. The rates, products, and activation parameters were all con- sistent with a mechanism involving the formation of an alkyl radical as one of the rate-determining steps. The steric crowding in 2-endonor- bornanecarbonyl peroxide appears to be significant. The rates and products from decomposition produced little or no evidence for partici-- pation infree radical reactions. 6.. Several new compounds were preparedin connection with the peroxide work. These were: 2-endonorbornyl 2-endonorbornanecarboxylate, 2-exonorbornyl 2-exonorbornanecarboxylate, 5-endonorbornenyl S-endonor- bornenecarboxylate, 5-exonorbornenyl 5-exonorbornenecarboxylate, and the gamma-3 lactone of 2-exochloro-3-endohydroxy-5-endonorbornane— carboxylic acid. 7. Some miscellaneous experiments involving attempted preparations of 7-norbornenecarboxylic acids and the identification of cyc10propyl- carbinyl cyclopropylacetate as the ester produced from the decomposition of cyclopropylacetyl peroxide were described. ‘1 10. 11 12. 13. 14. 15. 16.» LIT ERATU RE CIT ED » A. Todd, Perspectives in Organic Chemistry, Interscience Publishers Inc., New York, 1956,. pp. 265-314.. .- E. S. Gould,’ Mechanism and Structure in Organic Chemistry, Henry Holt and‘Company, New York, 1959, pp. 594-599. . J. Meinwald, Rec. Chem. Prog., 22, (1), 39 (1961)° .. J. D.. Roberts, W. Bennett, and R. Armstrong, J. Am. Chem. Sgc., 23, 3329 (1950); J. D. Roberts and W- Bennet, ibid., 22, 4623 (1954). . J. D.. Roberts, C- C. Lee, and W. H. Saunders, Jr., J. Am- Chem. Soc., 16, 4501 (1954). ..J. A. Berson, C. J. Olsen, and J. S. Walia, J. Am. Chem. Soc., «8’2, 5000 (1960) . C. Walling and R. B. Hodgdon, Jr., J. Am. Chem. Soc., 0, 228 M (1958). . H. S. Shine and D..M. Hoffman, J. Am. Chem. Soc., 83, 2782 (1961). .. C. Walling, Free Radicals in Solution, John Wiley and Sons, Inc. , New York, 1957, pp. 474-503. A. V. Tobolsky and R. B.. Mesrobian, Organic Peroxides, Inter- science Publishers Inc. ,, New York, 1954, pp. 72-87. .» J. Hine, Physical Organic Chemistry, McGraw-Hill Book Company Inc.,. New York, 1956,. pp. 412-419. E. 5. Could, 1_o_(_:. c_i_t_., pp. 714-720. W. E. Cass, J. Am. Chem. Soc., 653, 1976 (1946). K- Nozaki and P. D. Bartlett, J. Am. Chem. Soc. , 68, 1686 (1946). J. E. Leffler, J. Am..Chem. Soc., 2,2,, 67 (1950). M. Szwarc, Chem. Rev., 47, 75 (1950). 114 17. 18. 19. 20. 21. 22. 23. 24. 25. 115 G. 5. Hammond and L- M. Soffer, J. Am. Chem. Soc., 13, 4711 (1950). M. C- Ford and D.. Mackay, J. Chem. Soc. , 4620 (1957). M. C. Ford and D.. Mackay, J. Chem. Soc. , 1290 (1958). D. M. Teller, Doctoral Thesis, Michigan State University, 1957. R. Rembaum and M..Szwarc, J. Chem. Phys. ’22, 909 (1955). D- F. DeTar and C. Weis, J. Am. Chem.‘ Soc., 28, 4296 (1956). P. D. Bartlett and J. E. Leffler, J. Am. Chem. Soc., 13:: 3030 (1950). C. Walling, lo_c. £i_t_., p. 478. G. S..Hammond,. J. T. Rudesill, and F. J. Modic, J. Am. Chem. Soc., 73, 3929 (1951). 26. D. B. Denney and G. Feig, J. Am. Chem. Soc., 81, 5322 (1959). 27. 28. 29. 30. 31.. 32.. 33. 34. 35.- J.- E. Leffler and C. C. Petropoulos, J. Am. Chem. Soc., 22, 3068 (1957). D. B. Denney, J. Am. Chem. Soc., 2A8, 590 (1956). P. D. Bartlett and F. D. Greene, J. Am. Chem. Soc., 16, 1088 (1954). G. 8.. Hammond, J. H. Sen, and C.- E- Boozer, J. Amg..Chem. Soc., 13, 3244 (1955). J. Smid,-A. Rembaum, and M. Szwarc, J. Am. Chem.- Soc., 18, 3115 (1956). H. H. Lau and H- Hart, J. Am. Chem. Soc. , A83, 4897 (1959). M. S. Kharasch, J. Kuderna, andW. Nudenberg, J. Or r.g Chem. 19,1283(1955). F. D. Greene, J. Am. Chem. Soc., 21' 4869 (1955). D... F. DeTar and C. Weis, J. Am. Chem. Soc., lg, 3045 (1957). 36. 37. 38. 39. 40. 41.. 42. 43.- 44. 45. 46. 47.: 48. 49.‘ 50. 51.. 52. 53. 116 L. S- *Silbert and D. Saver-n, J. Am. Chem. Soc., '8’}, 2364 (1959). L. 8. Gilbert and D. Swern,‘ Anal. Chem., 39, 385 (1958). K. Alder, G. Stein, M. Liebmann, and E. Rolland, Ann., 514,. 197 (1934). .H. Diels and K. Alder, Ann., 460, 117 (1928). W. R. Boehme, E. Schipper, W. G. Schzysf, and J. Nichols, ._J.‘Am.. Chem-Soc.,,gg, 5488 (1958). J. D. Roberts, E. R. Trumbull, Jr., W. Bennett, and R. Armstrong, .J. Am. Chem. Soc., ’12,, 3116 (1950). D. Ver Nooyand C. S. Rondestvedt, Jr. , J. Am. Chem. Soc. , 17’, 3583 (1955). D. Ver Nooy and C. s. Rondestvedt, Jr., J. Am.- Chem. Soc., 76, 2315(1954L J. A. Berson and D. A. Ben-Efraim, J. Am. Chem. SOC°5§01’ 4083 (1959). ‘ ‘ K. Alder, K. Heimbach, and R- Reubke, Chem. Ber., 21, 1516 (1958). K. Alder, G. Stein, E. Rolland, and G. Schulze, Ann., 514, 211 D. P. Wyman, Doctoral Thesis, Michigan State University, 1957. K. Alder and H. F. Rickert, Ann., 543, 1 (1939). D. E. Applequist and G. F. Fanta, J. Am- Chem. Soc., "8‘2, 6393 (1960). W. von‘E. Doering and T. C. Aschner, J. Am.» Chem. Soc., 71, 838 (1949). 'w H. A. Bruson and T. W. Riener, J. Am. Chem. Soc.,lég, 723 (1945). E. A. Guggenheim, Phil- Mag., “2‘, 538 (1926). E. G. Foster, A. C. Cope, and F. Daniels, J. Am. Chem. Soc., 23' 1893(1947). 54. 55.. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66.- 117 R. Kh. Freidlina and Ye. I. Vasil'eva, DokladyAkad. Naugk. SSSR., 1’03, 85 (1955). H. Kwart and L. Kaplan, J.-Am. Chem. Soc., 26, 4072 (1954). T.‘M. Bruton and E. F. Degering, J. Am- Chem. Soc., ’63:, 227 (1940). R. R. Sauers, Chem. and Ind., 176 (1960). H. Gilman, J. A. Beel,- M. W. Bullock, G. E. Dunn, C. G. Brannen, and L. S. Miller, ' J. Am. Chem. Soc. , l1, 1499 (1949). ~D.‘ N. Kursanov, N. K. Baranetskaya, and V.-N. Setkina, Doklady Akad. Nauk. SSSR., 1.1.3,. 116 (1957). R. Riemschneider and M. Krl’iger, Monatsh. , ’92, 573 (1959). H. Hart and A. Holzschuh, Unpublished Results. R. A. Cipriani, Doctoral Thesis, Michigan State University, 1961. R. E. Pincock, Doctoral Thesis, Harvard University, 1959. W. COOper, J- Chem. Soc., 3106 (1951). M. C. Caserio, W- H. Graham, and J. D.. Roberts, J.-Am. Chem. Soc. preprint. M.‘ S. Kharasch, E. V- Jensen, and W- H. Urry, J.‘ Am. Chem. Soc., ’63, 1100 (1947), M. S. Kharasch, O.’ Reinmuth, and W. H. Urry, ibid., 1105 (1947). APPENDIX ' 118 119 Derivation of Equation Used in Calculating Rate Constants The equation used to follow the decomposition of the peroxides was obtained by the following manipulations. From the equation for a first order reaction one has / = fiche"kt (1) where [(C0) is an appropriate function of the initial peroxide concen- tration and [(C) is the same function at time t. If times t1, t2, t3, etc. , . and t1+A, tfi-A, t3+A, etc. , are selected where A is a constant increment, then the following equations are true: <7’> =1n[ .3358 :34 oudmmh O . mound-32-5 0&2”. oom . omH 00.. Qm o _ _ _ A 0H szemllliw 11: lam. 123 oo .3 so oosxosom . AGOEMHfiHV HenaonumCocmsuonuocoxmtm mo comfimomgooofl 0%. HOW worm EAoAComwd-O .mv 0.3.me moudcflz cw 08TH. om ON CA _ J _ 1.58 1: x 2 4 n x .38 1: x 34... n maoqm H- o5. (V+1A- 1A) 01801 124 Table 16. Decomposition of 2- -Exnorbornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 9o Determined by Infrared Techniques. Sample Time/min. Absorbancy 1 0.0 0.5806 2 15.0 0.5329 3 30.0 0.4680 4 45.0 0.4076 5 65.0 0.3483 6 90.0 0.2871 7 146.7 0.2055 8 170.0 0.1858 9 203.3 0.1661 10 241.7 0.1505 Table 17. Guggenheim Data for the Decomposition of 2- Exonorbornane- carbonyl Peroxide at 53. 9o Determined by Infrared Techniques. Time/min. At At-l-A At - At+A Log10(At - At+A) 10 0.548 0.172 0.376 -0.4248 20 0.508 0.167 0.341 -0.4672 30 0.468 0.162 0.306 -0.5142 40 0.432 0.156 0.276 -0.5590 50 0.396 0.151 0.245 -0.6108 A = 190 minutes Per Cent T ransmis sion 100 75 50 25 125 (PART 1“) l J J l L 5.60 5.60 5.60 5.60 5.60 Wavelength in- Mic rons Figure 46. Quantitative Infrared Spectra of the Decomgosition of 2-Exonorbornanecarbonyl Peroxide at 53. 9 . Per Cent Transmission 100 75 50 25 126 (PART 2) l 'l l I I~ 5.60 5.60 5.60 5.60 5.60 Wavelength inMicrons Figure 46. Quantitative Infrared Spectra of the Decomppsition of Z-Exonorbornanecarbonyl Peroxide at 53. 9 . .00 .mm «m ogxouom anonumnvocmcuonuoaoxmlm mo coflwmomfiooofl 0:... he worm toEmH. mfimuo> >undnu0wn< .hw. shamrm mound“: Gm 08TH. 127 oom OS 2: o... o _ . _ -_ _ [c.o O O O .146 O V. q S m 0 m U 3 IA 0 1N6 . . Loo _ 5 b _ 128 0 .mm «0 0303000 H>conu000smauonuocoxmtm mo Goflwmomgoo0fi 03.3 no“ uofim_500€0mm50 .wv 0udmfih 0 000332 GM 05MB 89.38 ‘ on 3. on on S _. A a A fl 0 166 O TEE 9.2 x no; u x moi . ed 1. .88 0.2 x 3.4.. u 000.0 m W 4 . - V Lmdu N. «to 129 Table 18. Appearance of Acid from the Decomposition of 2- Exonorbornane- carbonyl Peroxide in Carbon Tetrachloride at 53. 9o Determined by Infrared Techniques W Sample Time/minutes Absorbancy 1 0.0 0.0510 2 15.0 0.0730 3 30.0 0.1021 4 45.0 0.1323 5 65.0 0.1640 6 90.0 0.1940 7 146.7 0.2097 8 170.0 0.2097 9 203.3 0.2129 10 241.7 0.2207 Table 19. Guggenheim Data for the Appearance of Acid from the Decompo- sition of 2- -Exonorbornanecarbonyl Peroxide in Carbon Tetrachloride at 53. 9o Determined by Infrared Techniques Time/minutes At At+A At+A- At Log10(At+A- At) 10 0.064 0.216 0.152 -O.8181 20 0.085 0.217 0.132 -0.8794 30 0.105 0.218 0.113 -O.9469 40 0.124 0.218 0.094 .. -1.0268 50 0.142 0.218 0.076 -1.1191 A = 190 minutes 130 .00 .mm 00 0303.05” H>Conuso0cmnuonuonoxmtm - mo cofiwmomEoo0Q 0:... Eoum 084 m0 00G0n00mm< 05 Hem uofinm 05:... mdwh0> >onmnn003< .ww 0p5mfim 000.952 ,5 05TH. com omH on: om c _ n _ a. _ Do N.o Aou'eqxosqv 131 400.3383 0 .mm «0 0pflnou0nm 1:03.80 10G0Guonuocoxmtm mo Gowfimomfioo0fl 05. 50.8 304 «o 0090902584 0:... mam worm Ew0€0mm50 .om 095mmh om 00220344 E 05TH Om on Ca _ _ 1. .25 N..2 x 2.; u. r .28 m.2 x 2.7 H 00040 _1. m7 (1V .. V‘l'lv) OISor-I 132 Table 20. Energy of Activation for the Decomposition of Z-Exonorbornane- carbonyl Peroxide in Carbon Tetrachloride Determined by Titrimetric Techniques = -= w T /°K ‘ T-l/OK"1 k/minutes‘l loglok Slope /°K Ea/kcal. / mole 339.05 2.9499x10-3 49.1x10‘3 -1.30936 . 327.05 3.0581x10-3 12.35110:3 -1.91186 ---6.23x1o3 28.5' 317.65 3.1486x10'3 2.813110"3 -2.55207 Table 21. Entropy of Activation for the Decomposition of Z-Exonorbornane- carbonyl Peroxide in Carbon Tetrachloride Determined by Titrimetric Techniques k/minutes‘l Logmk Ea/2.303RT Logms Average 215*. ato logws 328. 35 .K/e.u. 49.1x10'3 -1.30936 18.37768 17.06832 12.3x10’3 -l.91186 19.05209 17.14023 17.09082 17.49 2.81x10-3 -2.55207 19.61598 17.06391 «5v 1le "11111111111111