ABSTRACT PROTONATED CYCLOPROPANE FORMATION IN THE DEAMINATION 0F CYCLOPENTYLCARBINYL AMINE By Kenneth Edward Martin The purpose of this investigation was to determine whether protonated cyclopropane intermediates are formed in the deamination of cyc10penty1carbinyl amine. The method used involved first, a study of the classical reaction pathways for the cyclopentylcarbinyl cation, and second, observation of the isotope position rearrange- ments resulting from deamination of a number of deuterium labeled cyc10penty1carbiny1 amines using mass spectroscopy and NMR. The alcoholic fraction resulting from deamination of cyclopentyl- carbinyl amine consisted of 74.82 cyclohexanol, 17.6% ldmethylcyclo- pentanol, 4.9% cyclopentylcarbinol, and 2.7% Eggggfz—methylcyclo- pentanol. 1~methylcyclopentanol could be formed only by classical carbonium ion rearrangement. The lack of isotope position rearrangement occurring in the formation of cyclopentylcarbinol from cyclopentyl- aminomethane-1,1fid2 is inconsistent with an equilibrated protonated cyclopropane intermediate, and was interpreted as indicating that this material results from classical intermediates only. However, the isotope position rearrangements observed in the formation of trans:2~methylcyclopentanol from l-aminomethylcyclo- pentanol-lfd_are consistent with an equilibrated protonated 1 Kenneth Edward Martin 2 cyc10propane intermediate, but inconsistent with a classical 1,3 hydride shift. The possibility of successive 1,2 hydride shifts or equilibration of classical ions involved in the formation of this alcohol was eliminated by deamination of cyclohexyl and ldmethyl- cyclopentyl amines and supported by the isotope position rearrange- ment pattern observed in the deamination l-aminomethylcyclopen- tane-3,4fd2. For cyclohexanol, the small amount of isotope position rearrange- ments observed in the deamination of l-aminomethylcyclopentane-lfd are consistent with a protonated cyc10propane intermediate yet inconsistent with an alkyl shift mechanism. The possibility of 1,2 or 1,3 hydride shifts accounting for the observed isotope position rearrangement was eliminated by deamination of cyc10penty1amino- m.ethane-1,1-Ic_1_2 and l-aminomethylcyclopentane-3,Afde. Also deamination of cyclohexyl and l-methylcyclopentyl amines eliminate equilibration between classical ions as a possible explanation for the observed isotope position rearrangement. Since the isotope position rearrange- ments found in cyclohexanol resulting from cyclopentylaminomethane-l,l-g2 also eliminate equilibrated protonated cyc10propanes as intermediates in the formation of this product, a non-equilibrated protonated cyc10propane is invoked to explain these results. PROTONAIED CYCLOPROPANE FORMATION IN THE DEAMINAIION OF CYCLOPENTYLCARBINYL AMINE By Kenneth Edward Martin A THESIS Submitted to , Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1972 To my Parents 11 ACKNOWLEDGMENTS The author wishes to express his gratitude to all those persons and organizations who have contributed in any way to this effort. Special thanks are extended to Professor G. J. Karabatsos for his guidance and friendship, and to Emilie Martin for her patience and understanding. iii TABLE OF CONTENTS INTRODUW ION O O O O O O O O O O O O O O O O O O O O O O O O O I. Protonated Cyclopropane Formation in the Propyl System II. The Structure of Protonated Cyclopropanes . . . . . . . III. Protonated Cyclopropane Formation in Alkyl Substituted Propyl Systems . . . . . . . . . . . . . . . . . . . . IV. Solvent Dependency of Protonated Cyclopropane Formation V. Protonated Cyclopropane Formation from Mono-Cyclic Systems . . . . . . . . . . . . . . . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . . . . I. Deamination of Cyclopentylcarbinyl Amine . . . . . . . II. Deamination of Cyclopentylaminomethane-l,1fg_ . . . . . III. Deamination of l-aminomethylcycIOpentane-3,4fig_ IV. Deamination of l-aminomethylcyclOpentane-lfd_ ? . . . . SUMMARY AND CONCLUSIONS . . . . . . . . . . . .'. . . . . . . . EXPERIMENTAL O O O O O O O O O O O O O O O O O O O O O O O O 0 NMR Spectra . . . . . . . . . . . . . . . . . . . . . . Gas Chromatography . . . . . . . . . . . . . . . . . . Deamination Procedure . . . . . . . . . . . . . . . . . Preparation of Trimethylsilyl Ethers . . . . . . . . . Preparation of Cyclopentyl Cyanide . . . . . . . . . . Preparation of Cyclopentylcarbinyl Amine . . . . . . . Preparation of Trans-2-Methy1cyclopentanol . . . . . . Preparation of N-(l-methylcyclopentyl) Formamide . . . Preparation of 1-methylcyclopenty1 Amine . . . . . . . Preparation of Cyclopentylaminomethane-l,1-22 . . . . . Preparation of Cyclopentanecarboxylic Acid . . . . . . Preparation of Cyclopentylmethanol-l,1--£l_,2 . . . . . . . Preparation of D borane . . . . . . . . . . . . . . . . Preparation of A -cyclopentenol . . . . . . . . Preparation of Tris(triphenylphosphine)ch ororhodium (I). Preparation of Cyclopentanol-3,4fd_ . . . . . . . . . . Preparation of Cyc10penty1tosylate-3,4-d_ . . . . . . . Preparation of Cyclopentyl-3,4-51__2 Cyanide . . . . . . . Preparation of l-aminomethylcyclopentane-3,4-d Preparation of Cyclopentanecarboxylic Acid-3,zgg_ iv Page ON 12 16. 21 24 24 28 38 46 56 59 59 59 59 6O 61 61 62 62 63 63 64 64 65 65 66 66 67 67 68 68 TABLE OF CONTENTS — Continued Page Preparation of Cyclopentylcarbinol-3,4-151“2 . . . . . . . . 69 Preparation of Cyclopentanol-l-d_ . . . . . . . . . . . . 69 Preparation of Cyc10pentyltosylate-1fd_ . . . . . . . . . 69 Preparation of Cyclopentyl-l-d_Cyanide . . . . . . . . . 70 Preparation of l-aminomethylcyclopentane-l-g_ . . . . . . 70 Preparation of Cyclohexanol-lfid . . . . . . . . . . . . . 71 REFEENCES O O O O O O O O O O O O O O O O _ O O O O O O O O O O O 72 LIST OF TABLES TABLE 1. 10. 11. 12. 13. 14. 15. + 3H7 . . . . . . . . . . . . . . . Calculated Energies for C3117+ . . . . . . . . . . . . . . . Calculated Energies for C Yield of Methylcyclopropane from the Deamination of Isobutyl Amine in Various Solvents . . . . . . . . . . . . . . . . . Deuterium Content of Cyclopropane Obtained from Deamination of 1-aminopropane-3,3,3-g__3 in Various Solvents Alcoholic Products from Deamination of Cyclopentylcarbinyl Mine 0 O O O O O 0 O O O O O O O O O O O O O O O I O C O O Alcoholic Products from Deamination of Cyclopentylaminome- thane-l , 1.9.2 o o o o o o o o o o o o o o o o o o o o o o 0 Expected Isotope Distribution in Products from the Deamination of Cyclopentylaminomethane-l,1-d ‘2 O O O C I O 0 Mass Spectrum of the Trimethylsilyl Ether of Cyclopentyl- carbinol . . . . . . . . . . . . . . . . . . . . . . . . . Mass Spectrum of the Trimethylsilyl Ether of Cyclopentyl- Garb 11101-1 , l-dz o o o o o o o o o o o o o o o o o o o o o 0 Mass Spectrum of the Trimethylsilyl Ether of Cyclopentyl- carbinol Isolated from the Deamination of Cyclopentyl- aminomethane-l ’ 1—12 0 o o o o o o o o o o o o o o o o o o o Alcoholic Products from Deamination of 1-aminomethy1- CYClOPentane-3 ’4-12 0 o o o o o o o o o o o o o o o o o o 0 Expected Isotope Distribution in Products from the Deamination of 1-aminomethy1cyclopentane-3,4-d2 . . . . . . Mass Spectrum of the Trimethylsilyl Ether of Cyclopentyl- carb inc 1 O I O O O O I I O O O O O O O O O O O O O O O O 0 Mass Spectrum of the Trimethylsilyl Ether of l-amino- methy1cyc10pentane—3 ’4-12 0 o o o o o o o o o o o o o o o 0 Mass Spectrum of the Trimethylsilyl Ether of Cyclopentyl- carbinol isolated from Deamination of 1-aminomethy1cyclo- pentane-3 ,4-22 0 O O O O I O O O O O O O O O O O C O O O 0 vi Page 7 8 17 19 25 29 3O 31 31 32 41 41 43 43 44 LIST OF TABLES - Continued TABLE Page 16. Alcoholic Products from Deamination of 1-aminomethyl- cyclopentane-l-d_ . . . . . . . . . . . . . . . . . . . . . . 47 17. Expected Isotope Distribution in Products from Deamination Of l-am1n0methy1CYClOpentane-l’i o o o o o o o o o o o o o o 47 vii FIGURE 10. 11. 12. 13. 14. 15. 16. 17. LIST OF FIGURES Solvolysis of Cyclopropane . . . . . . . . . . . . . . . . . Equilibration of an Edge-Protonated Cyclopropane . . . . . . Equilibration of a Corner-protonated Cyclopropane . . . . . . Isotopic Distribution Expected from Edge and Corner- protonated CycloprOpanfa in Reaction of Aluminum Bromide with l-bromopropane-l- C . . . . . . . . . . . . . . . . . . Deamination of 3—methy1-1-aminobutane . . . . . . . . . . . . Relation of Isotopic Scrambling to Equilibration between Various Protonated Cyclopropanes . . . . . . . . . . . . . . Possible Mechanistic Schemes for Deamination of Cyclopentyl- c arb iny 1 Anine O O O O O O O O O O O O O O O O O O O O O O 0 Classical Mechanism for Deamination of Cyclohexyl Amine, Cyclopentylcarbinyl Amine, and l~methy1cyclopenty1 Amine . . Possible Mechanisms for Formation of Cyclopentylcarbinol-1,l- dQ from Cyclopentylaminomethane-l,l--‘<_i_2 . . . . . . . . . . . NMR Spectrum of Cyclohexanol . . . . . . . . . . . . . . . . NMR Spectrum of Cyclohexanol-‘d.2 from Deamination of Cyclo- pentylam1nomethane-1 , 1_g-2 o o o o o o o o o o o o o o o o o 0 NMR Spectrum of Trans-Z-methylcyclopentanol-g2 from Deamination of Cyclopentylaminomethane-l,l-g2 . . . . . . . . NMR Spectrum of Trans-Z-methylcyclopentanol-jgim2 from Deamination of Cyclopentylaminomethane-l,1-_d_2 . . . . . . . . NMR Spectrum of Trans-Z-methylcyclopentanol . . . . . . . . . Preparation of 1-aminomethy1cyclopentane-3,4--g_2 . . . . . . . NMR Spectrum of Cyclohexanoljd from Deamination of 1-aminomethy1cyclopentane-3,4féz. . . . . . . . . . . . . . . NMR Spectrum of Trans-Z—methylcyclopentanol-d_ from Deamination of 1-aminomethy1cyclopentane-3,4-__2 . . . . . . viii 10 11 15 18 26 27 33 35 36 37 39 40 42 45 . 48 LIST OF FIGURES - Continued FIGURES 18. 19. 20. 21. 22. 23. 24. NMR Spectrum of Methyl Region of Trans-Z-methylcyclo- pentanoleg_ from Deamination of l-aminomethylcyclopen- tane-3 ,4-23 0 o o o o o o o o o o o o o o o o o o o o o o 0 NMR Spectrum of Cyclohexanol-lfd_ . . . . . . . . . . . . . NMR Spectrum of Cyclohexanolfid_from Deamination of 1-8m1n0methy1cyclopentane‘l-i o o o o o o o o o o o o o o 0 Possible Mechanisms for Formation of Cyclohexanol from the Cyclopentylcarbinyl Cation . . . . . . . . . . . . . . . . NMR Spectrum of Trans-2-methy1cyclopentanolfid_from Deamination of l—aminomethy1cyclopentane-1fid_ . . . . . . . NMR Spectrum of Methyl Region of Trans-deethylcyclopen- tanolfd_from Deamination of 1—aminomethylcyclopentane-lfd_. Mechanism for Deamination of 1-aminomethylcyclopentane-lfd ix Page INTRODUCTION Reports of research concerning the detailed nature of carbonium ion intermediates have occupied a major portion of chemical literature for the past fifty years. Some of this work represents major break— throughs and classical experiments of our time, but the majority of these efforts have often seemed to be almost trivial further examples of already demonstrated effects. Yet, since conclusions based on a single, perhaps anamolous event have occasionally led chemists astray, these further bits of evidence are necessary to build the strong experimental foundations upon which modern carbonium ion theory is based. Through the efforts of literally thousands of chemists more is known about the intimate nature, structure, and behavior of carbonium ions than about any of the other types of reactive chemical inter- mediates. Yet despite this vast body of knowledge there are still many gray areas of misunderstanding that have eluded satisfactory and conclusive explanations. Perhaps the best known of these is the question of nonclassical ion formation in the norbornyl system. Though the volume of work in this field can only be described as monumental, the question of the existence of these intermediates is almost as hotly debated today as it was ten years ago. It is only in a simpler system that the existence of a species with bonding, delocalized, sigma electrons has been described as resting on "firm experimental ground "(1). The formation of this nonclassical intermediate, a protonated cyc10propane, in a cyclic system is the subject of this thesis. 1 2 Protonated Cyclopropane Formation in the Propyl System The concept of protonated cyc10propane intermediates originated in 1953 when Roberts and Halmann (2) discovered that 1—aminopropane- 1-14C yielded upon deamination 1-propanol-14C in which 8.5% of the original 14C was not at C-1. They assumed it to be at C-2 and pro- posed that 172 of the originally formed l-propyl cation rearranged to a protonated cyc10propane I. Between 1959 and 1962, Skell and Starter (3) deoxidated l-propanol, demonstrated that the reaction proceeds through carbonium ion inter- mediates, and isolated cyc10propane from the reaction mixture. The formation of cyc10pr0pane conclusively demonstrated the existence of some type of interaction between C-1 and C-3 of the propyl cation, which Skell originally postulated (3b) as a face-protonated cyc10pro- pane II. 51+ CH II In 1962 Reutov and Shatkina repeated Roberts' experiment and 3 found 8.5% of the 14C in the resulting l-propanol at C-3. No 14C was found at C-2. Similar results were obtained from the reaction of l-chloropropane-l-IAC and ZnClz/H01 (5). In the same year, Karabatsos and Orzech (6) deaminated l-aminOpropane-l,1,2,2-d4 and reported that 12% of the resulting l-propanol had undergone a nominal 1,3 hydride shift. Both Reutov and Karabatsos concluded that 1,3 hydride shifts and not protonated cyc10propane intermediates were responsible for the observed results. A major breakthrough in protonated cyc10propane chemistry occurred when, in a series of classic experiments, Baird and Aboderin (7) passed cyc10propane through D28 and found significant amounts of 04 deuterium incorporated into the cyc10propane ring. Examination of the 1-propanol products in this reaction revealed the following deuterium distribution, based on incorporation of one deuterium per molecule: C1, 0.38D; C 0.17D; C 0.46D. These results cannot be adequately 2’ 3’ explained through the usual carbonium ion reactions of a propyl cation. The mechanism suggested for this transformation (See Figure 1) involved edge-protonated cyc10propanes as the product forming intermediates. This edge-protonated species contains a proton, or deuteron, simul- taneously bonded to two carbons; it differs from a face-protonated cyc10propane, that contains a proton symmetrically bonded to all three carbons, and from a corner-protonated cyc10propane that contains pro- tons bonded to only a single carbon atom. A variety of experimental and theoretical data indicate that, in simple acyclic systems, the edge- protonated formulation is superior to the other two. This aspect T\\ /1 ‘Er- ‘ ~. ‘\ +"D ZS:- IH / c ‘~CH '*CH C 2'- 2 2-- 2 . .;j; 2 ‘u sou son sou CHzD-CHZ-CHZ-OS CHZD-CHZCHZ-OS + CH3-CHD-CH2-OS CHB-CHZ-CHD-OS D2504 Solvolysis of Cyclopropane Figure l_(7b) will be covered in greater detail later. In 1964 Baird (8) further revealed that the cyc10propane obtained from the deamination of 1-aminopropane-3,3,3fid_ consisted of 43% 3 C3H4D2 and 57% C3H3D3. Without using protonated cyc10propane inter- mediates the only possible pathway to cyc10propane-1d3 involved 1,3 deuteride shifts prior to ring closure. However, such shifts should account for only 12% C3H3D3 (6). Hence, to explain the large 57% C3H3D3 actually found, a mechanism incorporating protonated cyclo- propane intermediates was required. In 1965 Lee and coworkers again studied the deamination of z, 1-aminopropane-1—14C (9). Instead of finding 8% 1 C at C-3 of the resulting l-propanol as reported by Roberts and Reutov, Lee found 2% 5 14C at 0-2. In addition, deamination of l-tritio-l—aminopropane (10) CH -CH -CH -NH -————-—€> CH -CH -CH -OH 3 2‘T 2 2 1‘3 «‘2‘1‘2 100% 2% 2% 96% 14C 14C 14C 14C yielded l-propanol with 1.5% T at C—3 and 1.5% T at C—2. Since it had already been shown that under these conditions, the 2-propy1 cation does not rearrange to the l-prOpyl cation (6) Lee's results were best interpreted in terms of protonated cyc10propane intermediates. The case for protonated cyc10propane was further strengthened when Karabatsos, Orzech, and Meyerson (11) examined the deamination of l-aminopropanol-2,2-d2. was found to have the following deuterium distribution: The 1-propanol isolated from this reaction CH3-CD2-CH2-NH2-—€>»(C2H3D2)CH20H + CZHSCDZOH + (CZHAD)CHDOH 97.9% 1.2% 0.9% The migration of deuterium from C-2 to C-1 eliminated a mechanism based solely on 1,3 hydride shifts. The greater abundance of CZHSCDZOH over (C234D)DHCOH is inconsistent with a series of 1,2 shifts. Hence, a mechanism involving protonated cyc10propanes is required. This conclusion was supported by studying the deamination of 1-aminopropane-1,l-d2 (11) and by a reinvestigation of the deamination of l-aminopropane-l,l,2,2-—_d4 (11). Further reports of the deamination of 1—aminopropane-l-14C (12, 13), 1-aminopropane-l-13C (14), l—aminoprOpane-3,3,3-g3 (14, 15), 1-amin0pr0pane--2,2--d_2 (12, 14, 15), 1-aminopropane-—1,1—-§_2 (15) and 1-tritio—1—aminopropane (12) served to firmly establish that the 6 isotope position rearrangements observed in reactions of the 1-propyl cation are characteristic of protonated cyc10propane intermediates rather than any combination of 1,3 with 1,2 hydride shifts (14). In addition, observation of the same general scrambling patterns in the reaction of l-bromopropane-l-IBC (16, 17) l-aminopropane-LZ-g2 (17) and l--amin0propane-1,l-g2 (17) with aluminum bromide, the reaction of 1-chloropropane-1—14C with aluminum chloride (18) and the inter- action of 1-propanol-l,l--_c1_2 and ZnClZIHCl (l9) demonstrate that the formation of protonated cyc10propane is characteristic of the l-propyl cation and not restricted to ions produced by deamination reactions. It should be pointed out that protonated cyc10propane formation represents at best a minor pathway in the reactions of the l-propyl cation, accounting for 4%-6% (12, 14) of the product in deamination reactions. The Structure of Protonated Cyclopropanes Though the intermediary of protonated cyc10propanes has been firmly established, the question of their exact structural formulation has not been fully resolved. Both theoretical and experimental inves- tigations have attempted to determine whether a corner I, face II, or edge III, protonated structure best represents the actual inter- mediate. ,93‘3 2 + Ca? ~ ’I + \\\ H k\ + "H CH \ —-— c “ 'CH 2 _ CH2 CH2 CH2 2 2 7 In 1964 Hoffmann (20) used extended Huckel theory to calculate that the edge-protonated formulation gives the best picture of the actual intermediate. This conclusion was supported and extended by Petke and Whitten (21) who employed gbfinitio-SCF-MO calculations to find that an edge-protonated cyc10propane represented the lowest energy equilibrium geometry for C Further evidence in favor of the + 3H7 ' edge-protonated formulation was provided by Smith, Kollman, and coworkers (22), (23), who used modified CNDO and INDO methods of cal- culation. Their conclusions, summarized in Table 1, also indicate that edge-protonated cyc10propane represents the low energy form of + 03H7 . Table 1_(22), (23) Calculated Energies for 03117+ Relative energy in Kcal/mole Face-protonated cyc10propane 62 l-propyl cation 25 2-propy1 cation 0 Corner-protonated cyc10propane - 3 Edge-protonated cyc10propane -14 More recently, Dewar and Bodor (24) studied this problem by using the MINDO/2 method and established the following order of stability: face-protonated cyc10propane < n-propyl cation < corner-protonated cyc10propane < isopropyl cation < edge-protonated cyc10propane. The conclusions that an edge—protonated cyc10propane is the energy minimum for C3H7+ has been called into question by calculations (25) based on gbfinitio-SCF~MO theory with modifications that allowed exten— sive variations in geometry of the ions involved. Relative energies 8 found by this method based on the optimized geometry for each species are given in Table II. These results indicate that a corner-proton- Table 2 (25) + 3H7 Relative energies in Calculated Energies for C Kcal/mole 2-propy1 cation 0 l-propyl cation 16.9 Corner-protonated cyc10propane 17.3 Edge-protonated cyc10propane . 27.1 Face-protonated cyc10propane 139.6 ated species is the most stable of the cyclic C3117+ ions, and that both the l-propyl cation and 2-propy1 cation are more stable than any type of protonated cyc10propane. It should be pointed out that all of the above calculations are based on isolated molecules in the gas phase. Introduction of an energy of solvation factor could easily alter these results. Experimental investigations into the structure of protonated cyc10propanes have attempted to determine if the observed isotope scrambling due to this intermediate more closely fits that predicted by face, edge, or corner—protonated formulations. Experimental results agree with theoretical calculations in eliminating face-protonated cyc10propanes from consideration. Among many findings that support this conclusion was the discovery of a non-symmetrical incorporation of deuterium when cyc10propane was solvolyzed in D2304 (7). A face—protonated cyc10propane would have predicted incorporation of equal amounts of deuterium at C-1 and C-2. A distinction between edge and corner-protonated structures is j: 1.D2804 > CH3 - C}:2 - C;2 - OH 2.1120 T 0.46D 0.17D 0.38D more difficult since both formulations can account for either partial or total scrambling of isotopic labels. Furthermore, since equili— bration between two edge-protonated forms can be best pictured as preceding through a corner-protonated transition state (12), (14), (See Figure 2), and because equilibration between two corner-pro- tonated forms would necessitate an edge-protonated transition state (12) (See Figure 3), distinction between the two forms becomes a problem of determining which is the intermediate and which is the transition state (or higher energy intermediate). V: a + / U mt". F l CDZH-CHZ-CHZ-Y Equilibration of an Edge-protonated Cyclopropane Figure 2_(14) 10 k k k I’Té 4') 1+ .qu ‘1—9 etc. m’_ _CD<— <—-— ,'+-~. <——— 2-=——— 2 CH .. -CD H 2 2 k2 (Y) k2 (Y) CH3-CH2-CD2Y CHDz-CHZ-CHZ-Y + CHB-CDZ-CHZY Equilibration of a Corner-protonated CycloprOpane Figure 3_ A distinction between these two structures can be made, however, by examining reactions in which only partial equilibration of isotopic labels occurs (18) (19). In these cases, equilibration between the various edge-protonated or corner-protonated structures is competitive with the rate with which the intermediates collapse to products. Consequently, a greater amount of product can be expected to result from the first formed ion rather than any others in equilibrium with it. Hence, in the reaction of l-bromopropane-l-IBC with aluminum bromide (16), for example, a corner-protonated intermediate would give products with a greater amount of 13C at C-2 than at C-3 (See Figure 4), whereas an edge-protonated intermediate would predict the opposite. The actual product from this reaction contains nearly three times as much 13C at C-3 than at C-2. An edge—protonated structure is therefore 11 * CH3-CH2-CH2-Br R5 31“ (mfmfcuz-Br - * ,’ + \\ ————> e + l AlBr3 CH; :3 CH2 CH3-CH2-CH2-Br * + CH3-CH2-CH2 - * * AlBr4 CH2- “H Br- cna-cnz-cuz-Br \\ +I,’ ) * + CH2 411/2 CHa-CHZ-CHz-Br Isotopic Distribution Expected from Edge and Corner-proton- ated Cyclopropanes in Reaction of Aluminum Bromide with 13 l-bromopropane-l- C Figure-4 indicated as the product forming intermediate (16). * AlBr3 CH -CH -CH -Br -————————4’ CH -CH -CH -Br 3 2 2 3 2 2 MT 10.6% 3.7% 85.7% 13C 13C 13C This conclusion is supported by a variety of other experiments (7)- In fact, all reactions in which a distinction can be made between edge and corner-protonated cyc10propanes indicate that edge-protonated structures are the lower energy intermediates. Figure 2, therefore, gives the best mechanistic picture of reactions involving protonated cyc10propanes. It should be emphasized that a variety of isotOpic scrambling patterns can be expected from this type of intermediate. For example, 12 in Figure 2, if k1 > > k2, any isotopic label will be symmetrically scrambled over all possible positions. The deamination of 1-propy1 amines (9) (10) fall into this category. However, if k2 > > k1, then products will appear to have been derived from a simple 1,3 shift. The reaction of 1-pr0panol with ZnC12/HC1 (19) is an example of such a case. Since the bulk of experimental and theoretical evidence favors the edge-protonated formulation, it will be used through the remainder of this thesis. However, it should be noted that all of the above experiments dealt with acyclic substrates. In view of the relatively small energy difference between edge and corner-protonated species, the various stereochemical constraints imposed by cyclic and poly- cyclic structures could alter the relative stabilities of these two ions. Protonated Cyclopropane Formation in Alkyl Substituted Propyl Systems As in the propyl system, evidence for protonated cyc10propane intermediates in more highly substituted systems is based on cyclo- propane formation and on isotope position rearrangement characteristic of protonated cyc10propane structures. However, the role protonated cyc10propanes play in the reactions of these larger molecules has been found to be sharply reduced. Two explanations have been advanced (26, 27, 28) to explain this phenomenon. First, the introduction of larger groups onto a protonated cyc10propane ring would tend to in- hibit the formation of this intermediate because of unfavorable l3 1, 2-eclipsing interactions. And second, other rearrangement paths CH3 ‘\ ‘ s~11 CH - CH ‘-————-—-€D I , CH/ 2 I’I 3 leading to relatively stable ions are usually available. ~11 . 4- CH3 /-——) CH3-("J-CH3 \\ . CH 4. CH + CH-CH 2 ~01! CH 3 3 CH - Protonated cyc10propane intermediates have, however, been de- tected in several systems larger than propyl. For example, in 1965 Friedman and coworkers (29) isolated methylcyclopropane from the deamination of isobutyl amine and based on this evidence, postulated an edge-protonated cyc10propane as its precursor. The possibility CH3 + “3‘41 - '" \ H ‘ +.’ H /cn - CHz——-> \ 36H ————> CH3 CH//' 2 3 of carbenoid intermediates was eliminated by studying the deamination of isobutyl amine-1,1-d2 (30). That protonated cyc10propanes are, however, less important in this system was shown by Skell and Starter (3b), who found that introduction of a single methyl group at C-2 of l-prOpanol decreased cyc10propane formation upon deoxidation of the resulting 2-methy1—1—propanol by a factor of approximately 2.5 compared to l-propanol. In an extensive study of the isotope position rearrangement occurring in the aqueous acid deamination of isobutyl amine-1,1551”2 14 (26, 31), Karabatsos and coworkers detected no protonated cyc10propane intermediates involved in the formation of alcoholic products. Based on the formation of methyl cyc10propane, however, protonated cyclo- propanes were estimated to account for 0.6% of the products derived from this reaction. In comparison, the yield of protonated cyc10pro- pane derived products from l-propyl amine under similar deamination conditions was found to be 6% (14). It should be pointed out that protonated cyc10propane formation in this system has been found to be highly solvent dependent (32, 33), with higher amounts of methyl cyc10propane formed in solvents of low polarity. In a related system, Skell and Maxwell (34) isolated ethylcyclo— propane and 1,2-dimethylcyclopropane in 3.1% and 2% respective yields from the deoxidation of 2—methyl-l-butanol. Results from studies of the n-butyl cation paralleled those of the isobutyl system. Initial discovery of l-methylcyclopropane in the deoxidation of l-butanol (3b) was followed by confirmation of cyc10propane formation in the deamination of l-butyl amine (27, 28, 29). But again, extensive studies (28) of the deamination of l-butyl- 1,1fid2 amine, 1-butyl-2,2--£i__2 amine, and l-buty1-3,3--g_2 amine produced no evidence for protonated cyc10propane precursors to substitution products. As before, addition of a single methyl group to the prOpyl system reduced the role of protonated cyc10propanes in the reaction mechanism by an approximate factor of 10. Deamination (35, 36) and solvolysis (36) studies of deuterated l-pentyl and l-hexyl amines and tosylates indicate that straight chain compounds longer than four carbons have little or no tendency to form protonated cyc10propane intermediates in these reactions. 15 Deamination (37) of neopentyl-l—13C amine, and neopentyl--1,l--‘d_.2 amine as well as solvolysis (37) of the corresponding tosylates indi- cate no protonated cyc10propane precursors to substitution products. In addition, no cyclopropyl compounds were isolated from deamination of neopentyl amine (3b, 37) deoxidation of neopentyl alcohol (3b, 39) or solvolysis of neopentyl tosylate (38). The absence of products derived from a geminal, dimethyl substi- tuted protonated cyc10propane was observed in the deamination of 3-methyl-l-buty1 amine (40, 41). 1,2 dimethylcyclopropane was isolated as 1.5% of the product, and since no 1,l-dimethylcycloprOpane, the expected product from direct formation of a protonated cyc10propane by the primary cation, was found, a mechanism using an initial 1,2 hydride shift followed by protonated cyc10propane formation from the secondary ion was suggested (See Figure 5). ([113 (H3 + H3C—CH-CH2-CH2-NH2—>CH3-CH-CH2-CH2 7949 " H CH 3+ CH3-CH-CH-CH3 '———5P Deamination of 3-methy1-l-aminobutane Figure.§ In agreement with this mechanism, the deamination of 3—methy1—2- aminobutane (40, 41) gave a 15% yield of l,2-dimethylcyclopropane. Despite the large amount of cyc10propane formed in the above 16 reaction, most secondary carbonium ions form smaller amounts of pro- tonated cyc10propanes than do their less stable primary counterparts. For example, while methyl cyc10propane can be isolated as 2% of the hydrocarbon products from the deoxidation of l-butanol, 2-butanol gives only a trace ( :— s T + TI 0%" O/ "—9 0A?— 1%6‘0/ mg (m Possible Mechanistic Schemes for Deamination of Cyclopentylcarbinyl Amine Figure.1 intervenes in the rearrangement sequence. First, classical reaction pathways were explored by deamination of cyclohexyl and l-methyl— cyclopentyl amines, allowing entrance into the above mechanistic scheme at the corresponding carbonium ions. Second, deamination of several deuterated cyclopentylcarbinyl amines were studied in order to determine if the resulting isotope position rearrangement patterns are more consistent with classical or non-classical intermediates. Cyclohexyl amine was deaminated under conditions identical to those used for cyclopentylcarbinyl amine, and the resulting material was analyzed by gas chromatography. The only alcohol found was cyclo- hexanol. l-methylcyclopentyl amine was deaminated under conditions identical to those used for cyclopentylcarbinyl amine, and the 27 resulting material was analyzed by gas chromatography. The only alcohol found was l-methylcyclopentanol, These results allow the 13ft..hand portion of Figure 7 to be modified as follows: v //” me o > VIII Classical Mechanism for Deamination of Cyclohexyl Amine, Cyclopentylcarbinyl Amine, and l-methylcyclopentyl Amine Figureig These changes simplify a mechanistic interpretation of this re- action in several ways. First, the possibility of an equilibrium interchange between the three ions VI, VII, and VIII has been elim- inated. Second, the possibility of Ix arising from VII by a series of 1,2 hydride shifts has also been eliminated. This conclusion is strongly supported by the 100% trans stereochemistry of IX. Also, 28 because of this stereospecificity, a classical 1,3 hydride shift giving rise to IX would require attack of solvent to be concerted, or very nearly so, with hydride transfer. The formation of IX from a protonated cyc10propane carries no such restrictions. The most widely accepted type of evidence for protonated cyclo- propane intervention is the observation of isotope position rearrange- ment patterns typical of this type of intermediate and not consistent with a classical mechanism. Hence, in order to examine the possibil- ity of protonated cyc10propane participation in this rearrangement, the following deuterated cyclopentylcarbinyl amines were prepared and deaminated: Deaminationg;Cyclopentylaminomethane-l,l-d2 Cyclopentylaminomethane-l,l-_c1_2 was prepared by reduction of cyclopentylnitrile with lithium aluminum deuteride and deaminated under conditions identical to those used for cyclopentylcarbinyl amine. Analysis of the resulting alcohols indicated a product dis- tribution as in Table 6. Each of the products, with the exception of l-methylcyclo- pentanol which could only arise by a classical 1,2 hydride shift, was further analyzed to determine if the deuterium distribution indicated the presence of protonated cyc10propane intermediates. Table 7 shows the expected deuterium distribution pattern arising 29 Table.§ Alcoholic Products from Deamination of Cyclopentylamino- methane-1,1-d —2 on G 74.8% H3 O< 17 . 8% on O/\°H 4.8% on or 3 (Based on Alcoholic Fraction - 100%) from three types of intermediates: (A) a path involving only classical alkyl and hydride shifts, (B) a route in which products are derived from a non-equilibrated protonated cyc10propane (the first non-classical intermediate capable of giving a particular product is the only intermediate involved in the formation of that material), and (C) a mechanism depending only on an equilibrated protonated cyc10propane. The mass spectrum (See Table 8) of the trimethylsilyl ether of cyclopentylcarbinol-jg2 isolated from the reaction mixture was compared to the spectra of authentic cyclopentylcarbinol-1,lfgz, and of non- deuterated cyclopentylcarbinol (See Tables 9 and 10). These results indicate that cyclopentylaminomethane-l,lfd_ showed no loss of 2 deuterium from the one position during deamination. 30 Table 1 Expected Isotope Distribution in Products from the Deamination of Cyclopentylaminomethane-l,1-11”2 Product OH CH OH 3 Mechanism M‘QH on D Classical CHDZ * Shifts Only D on 2 ‘- cm Non-Equilibrated on D CHDZ Protonated D 0H <:::Z"‘ Cyclopropane 2 ‘ ‘OH H Equilibrated on m D D “”2 D on 2 Protonated H D . Cyclopropane 2 D “OH ‘OH 2 $2 @011 (fog *Product from a classical 1,3 hydride shift. Q CH D Q: 39"” D ‘1) \OH ‘OH Though this is perfectly consistent with a classical mechanism involving capture of the primary carbonium ion by solvent, it does not in itself rule out participation of a protonated cyc10propane in the formation of this product. The possibility exists that the pro- tonated cyc10propane did indeed form, but was attacked by solvent before any isotOpe position rearrangement could occur (a non-equili- brated protonated cyc10propane), as shown in Figure 9. However, in view of work cited earlier (15, 30, 31) concerning solvent effects 31 Table‘g Mass Spectrum of the Trimethylsilyl Ether of Cyclopentylcarbinol M/e Pk. Ht. Mono. 107 2.7 1.5 106 4.1 0.4 105 44.4 31.0 7.8% of Mono. 2 99-107 104 35.0 2.8 0.7% " 103 335.0 334.2 83.1% " 102 2.7 1.4 101 12.1 11.3 + 100 2.0 0.3 M/e 103 = (p-C5H9) 99 15.4 15.4 Table.g Mass Spectrum of the Trimethylsilyl Ether of * Cyclopentylcarbinol-1,1--g_2 M/e Pk. Ht. Mogo. 108 3.9 0.4 107 42.5 28.7 6.7% of Mono. 2 99-108 106 37.2 3.0 0.7% " 105 350.0 347.7 80.6% " 104 21.0 20.0 4.4% " 103 9.4 8.7 102 5.2 4.3 101 8.4 7.7 100 5.3 4.8 99 6.2 6.2 * Parent ion analysis of this compound gave an isotopic distribution of 95.4% d., 4.6% 91 32 Table'lg‘ Mass Spectrum of the Trimethylsilyl Ether of Cyclopentyl- carbinol Isolated from the Deamination of Cyclopentyl- * aminomethane-—1,l--~.d__2 M/e Pk. Ht. Mono. 108 1.4 0.1 107 15.9 10.8 6.5% Mono. 2 99-108 106 14.0 1.3 0.8% " 105 129.0 127.5 77.0% " 104 11.3 10.8 6.5% " 103 5.0 4.3 136% " 102 3.1 2.3 101 4.6 4.0 100 2.3 1.8 99 2.7 2.7 Parent ion analysis of this compound gave an isotopic distribution of 93.0: $2, 6.9% $1, 0.1: 90 33 and protonated cyc10propane, such a possibility should be judged as unlikely, since maximum isotope position rearrangement occurs in highly polar solvents such as the H20/HClO4 system used. Furthermore, solvent attack on an ion such as (X) would be expected at the secondary carbon, yielding ££§n§72—methylcyclopentenol, rather than at the primary position. Hence, it is unlikely that a protonated cyclo- prOpane intermediate is involved in the formation of this primary alcohol. Possible Mechanisms for Formation of Cyclopentylcarbinol-l,1--_d_2 from Cyclopentylaminomethane—l,l-_d_2 Figure 2_ A search for indications of protonated cyc10propane intermediates in the formation of cyclohexanol also proved fruitless. While a classical alkyl shift would result only in cyclohexanol-2,2—d2, a mechanism involving an equilibrated protonated cyc10propane would yield several deuterated cyclohexanols, including a species with a deuterium substituted in the one position. A careful comparison of 34 the NMR spectra (See Figures 10 and 11) obtained from cyclohexanol and cyclohexanol isolated from the deamination mixture indicated no incorporation of deuterium into the one position. However, due to the large percentage of cyclohexanol formation in this reaction, a small contribution from an equilibrated protonated cyc10propane intermediate could have been masked. And again, the absence of any isotope position rearranged product does not necessarily eliminate the possibility of protonated cyc10propane intermediates, but only of equilibrating protonated cyc10propanes. In view of experiments cited earlier (50) involving deamination of labeled cyclopentylcarbinyl amine, approximately 3%—9% of the intermediate cyclohexyl cation should have undergone a 1,2 hydride shift, resulting in a substantial amount of deuterium at C-1 in the product alcohol. That this was not observed indicates hydride shifts in the cyclohexyl cation under present conditions occur to either a considerably smaller extent than that reported, or not at all. The significance of this fact will be more fully explored later. Because of its 100% trans stereochemistry, 2-methy1cyclopentanol is the most likely product to examine for evidence of protonated cycloprOpane involvement. Though both a concerted 1,3 hydride shift and a non-equilibrated protonated cyc10propane intermediate give the same product, equilibration between various protonated cyc10propane forms will result in loss of deuterium from the methyl group and its accumulation at C-1 and C-2 of the ring. Deuterium at C—2 should be detectable by observation of an NMR splitting pattern for the methyl group consistent with a vicinal deuterium. Unfortunately the spectrum (See Figure 12) of this product does 35 H N m s m we age . p F L \ \ . \ \ _ 1 1 a a \ \ a \ \ a g}. , €J¢\/\ was com a suoHB amoam mfizmavsm we mm + mmraosousoeuaohoamnuoanmlmcouu H: om NI: via.HIosmnumaocaawaeusoeoaoko mo soHumswamoo Boom mmrHosmuaoeoHozoahnumanmumsmue NH: musmHm 40 m ma Bee \P 11V \. a .a\ \A| . Dismay ooHuem amHum> db \ P H;- db ~ ----... -_—.. ...- meo com a nuows emo3m mHzmovsm ms ms + ow..--.-—_——— —... -H-.H‘- -... Hocmusoeoflomoamzuoalmlmamuu H: ow HonousoeoaohoHanumalmlmcmua .mw musmHm h) R Q Eé 9 (Based on Alcoholic 41 Tablelll 1-aminomethylcyclopentane-3,4-d 74.8% 17.4% Table'lg Fraction - 100%) Alcoholic Products from Deamination of Expected Isotope Distribution in Products from the Deamination of l-aminomethylcyclopentane-3,4--‘d_2 Products ‘0. O’ OH Mechanism Cf . ‘ OH Classical on D D CH3 * Shifts Only 0" D D “OH Non-Equilibrated H CH Protonated Di:§::3r/\\OH D S 7 3 Cyclopropane .D D “OH . D Equilibrated H H3 H a 3 :0” ma C2311 ”12’ W” Cyclopropane D ‘NOH “OH D D *Product from a classical 1,3 hydride shift. 42 H202 'OH 5 NaOH _— H20 2 1( MC Po) ’,T3 3 3 I tosyl-Cl D mno com a :uoHS emmzm Inseam ma 8 + N.muHosssseoH98 H; 3 mmre.miosmuaoeoaomoHmsuoaocfiamla W\ mo soHuosHamoa aoum mmraosoxonoaomo NH. 3&2 C 46 These results do, however, support the earlier conclusion that the relatively large amounts of 1,2 hydride shifts previously ob- served in these systems (50) are much less important under present conditions. As expected, the NMR spectra (See Figures (17 and 18) of trans: 2-methylcyclopentanol isolated from the reaction mixture showed no deuterium at either C—2 of the ring or in the methyl group. This indicates the absence of any significant amount of successive 1,2 or 1,3 hydride shifts which would interfere in the analysis of isotope position rearrangements in the formation of this product from other deuterated cyclopentylcarbinyl amines. Deamination of l-aminomethylcyclopentane-l-d l-aminomethylcyclopentane-ljd_was synthesized by treatment of cyclopentanone with lithium aluminum deuteride, conversion of the resulting alcohol to the tosylate, reaction of the tosylate with sodium cyanide, and reduction of the nitrile to the desired amine. Deamination was accomplished under the same conditions as before, and the resulting alcohols were isolated by gas chromatography. The expected isotope distribution in products arising from three possible mechanisms is given in Table 17. Comparison of the NMR spectra (See Figures 10, 19, 20) of cyclo- hexanol and authentic cyclohexanol-lfd_with that of cyclohexanol isolated from the reaction indicated a 10% loss of deuterium from C-l. Since it has already been shown that under present conditions, 1,2 hydride shifts could not account for such a large amount of 47 Table 16_ (Based on Alcohols = Table'lz Alcoholic Products from Deamination of l-aminomethylcyclopentane-l-d ‘0. O< 74.7% 17.7% 4.8% 2.9% 100%) Expected Isotope Distribution in Products from Deamination Mechanism Product of l-aminomethylcyclopentane-l-§_ Classical Shifts Only Non-Equilibrated Protonated Cyclopropane “OH EQUilibrated D 01301101581 3 CH3 Protonated H H4 on ‘D <::Z:D Cyclopropane 0 OH OH 1 CT” creme D on D \OH *Product from a classical 1,3 hydride shift. is; 48 N m H» Sam 22;: meo 00H u :uoHs eoosm sonoM Hazuox 2 s «h- q- C!!- oc\em sowuo> one can u euoHs emozm umre.mioemusoeoaozoaanuoaocHamiH mo aoaumswamon scum _mrHosousoeoaohoaznuoatmimomma e .Mw ouswfim ooHi meo ooH u snows eooam 9 £585. 9.. 8 + A. mmIHosmusoeoaohoHanuoalmlmomuh mo H: oq one.muosmusoeoaomoHanuoaoswaola mo sowumsfiamoo aoum mmiaoceanoeoaomoamsuoalmlmsmue Ho 83% H.332 Mm 3am:— 50 ._o H ‘ ~-- .22?” nub oo\om gang, one can a H333 30% filaiaosoxonoaoho m ma 5&4 db see 51 N m .V 8am 28> C©\0m. GGHHNNV 4... .— «h- «r- . 30 com I 533 326 filalmcmuawaoaohuHmsumaoaaamna mo sown—madame Baum .wlfloamxmaoaoho mm Mmmmfilm 52 rearrangement, a protonated cyc10propane intermediate must be invoked to explain these results. Since no cyclohexanol with deuterium at C-1 was isolated from the deamination of cyclopentylaminomethane-l,ljg_, the protonated cycloprOpane intermediate that leads to cyclohexanol in this reaction must be non-equilibrated, i333, cyclohexanol must arise from.the first nonclassical intermediate capable of giving this product, and not from any other intermediate resulting from equilibrating pro- tonated cyc10propanes. It should be pointed out that protonated cyc10propanes could arise from two sources in this reaction as shown in Figure 21. Possible Mechanisms for Formation of Cyclohexanol from the Cyclopentylcarbinyl Cation Figure.gl The data of Edwards and Lesage (49) could be interpreted as if path 3 accounted for approximately 2% of the cyclohexyl cation with respect to elimination reactions. Furthermore, it is logical to 53 expect that ion XII is more stable than ion XI, since XII carries a positive charge dispersed over two secondary carbons while XI has the charge spread over one secondary and one primary carbon. Hence, a considerable portion of the first formed protonated cyclo- propane, XI , would be expected to rearrange to XII, and give cyclohexanol as product. Therefore, a relatively large amount of protonated cyc10propane intervention in the formation of cyclohexanol is not unreasonable. More conclusive evidence for the existence of protonated cyc10propane intermediates in this reaction results from examination of 2~methylcyclopentanolfd_isolated from this product mixture. As shown in Table 17 only an equilibrated protonated cyc10propane could yield a product mixture containing deuterium at both C-1 and C-2 of the cyclopentyl ring. The NMR (See Figures 22 and 23) of 552227 2~methy1cyclopentanolfg_clearly shows the methyl region as a triplet (J = 0.9) resulting from methyl-deuterium coupling, superimposed on a doublet (J = 6.0) resulting from methyl-proton coupling. Hence an equilibrating protonated cyc10propane must be an intermediate in the formation of this alcohol from cyclopentylcarbinyl amine. 54 «b ——.—. ‘~O--—... ' M .. —..-..-.— - p- . d- ‘4» 03 on swab; one can u pupas madam .mralmamunmaoau%oahsumaoawamua mo coaumofiawmn aoum .mrHoomuamnoaohoahnuoalmlmcmue mm 3&2 m c and 55 oo\mm amfium> one on u nuufia madam .mranmoouommoaohoahsumaoawawna mo coaumofiamon aoum .mraoqmuaoaofiomoamnuoalmmemmm mo cowwom ahsumz mIN. 3m»; Summary and Conclusions The foregoing results conclusively demonstrate that protonated cyc10pr0pane intermediates are involved in the formation of cyclo~ hexanol and trans:2~methylcyclopentanol in the deamination of cyclo- pentylcarbinyl amine. This conclusion was reached in the following manner. The isotope position rearrangements observed in the formation of Eggggfz-methylcyclopentanol from.l—aminomethylcyclopentanol-lfg_ are consistent with an equilibrated protonated cycloprOpane inter- mediate, but inconsistent with a classical 1,3 hydride shift. The possibility of successive 1,2 hydride shifts or equilibration of classical ions involved in the formation of this alcohol was eliminated by deamination of cyclohexyl and 1~methylcyclopenty1 amines and supported by the isotope position rearrangement pattern observed in the deamination l-aminomethylcyclopentans-3,41g2. For cyclohexanol, the isotope position rearrangements observed in the deamination of l-aminomethylcyclopentane-lfg_are consistent with a protonated cyc10propane intermediate yet inconsistent with an alkyl shift mechanism. The possibility of 1,2 or 1,3 hydride shifts accounting for the observed results were eliminated by deamination of cyclopentylaminomethane-l,1-g2 and l-aminomethylcyclopentane-3,4fg2. The isotope position rearrangements found in cyclohexanol resulting from cyclopentylaminomet:hane-1,l-—g_2 also eliminate equilibrated protonated cyc10propanes as intermediates in the formation of this product. Deamination of cyclohexyl and l-methylcyclopentyl amines eliminate equilibration between classical ions as a possible 56 57 explanation for the observed isotope position rearrangement. A mechanistic scheme, shown in Figure 24, can thus be written that is consistent with the above observations. Mechanism for Deamination of l-aminomethylcyclopentane—ljd Figure.g§ The above outline shows l-aminomethylcyc10pentane-lfd as the starting amine. Similar sequences could be drawn for the other cyclopentylcarbinyl amines discussed in this thesis. Also, proton- ated cyc10pr0pane equilibration sequences slightly different than that shown, yet also consistent with the data, could be drawn. Any such sequences, however, must account for the non-equilibrated 58 nature of the protonated cyc10propane precursor to cyclohexanol. For example, an ion such as XIV resulting from cyclopentylamino- methane-1,111”2 would yield only cyclohexanol-2,2»g2 and give no alcohol with deuterium at C-1, as observed. It is difficult to estimate the exact amount of protonated cyc10propane formation in this reaction. Because of the large amount of isotope rearrangement found in ££§n§72~methylcyclopentane (See Figures 22 and 23) and because of the 1001 ££§g§_stereo- chemistry of this product, it is assumed that all of this compound comes from protonated cyc10propanes. Though classical shifts are undoubtedly responsible for the bulk of cyclohexanol formation, the surprisingly large amount of isotope position rearranged product isolated from l-aminomethylcyclopentane-l-g, coupled with the nec- essary occurrence of a protonated cyc10propane capable of giving cyclohexanol in the equilibrating system necessary to explain the isotope position rearrangement in the formation of £522§72~methy1- cyclopentanol, indicate that a substantial amount, at least 32-101, of cyclohexanol formed in this reaction results from protonated cyc10propanes. Consequently, a minimum of approximately 5% of the total pro- ducts derived from the deamination of cyclopentylcarbinyl amine can be said to result from protonated cycloprOpane. This large amount of nonclassical ion formation in competition with classical pathways that result in relatively stable secondary and tertiary ions fur- nishes ample support for the hypothesis that protonated cyc10propane formation is enhanced in sterically favorable cyclic systems. EXPERIMENTAL NMR Spectra. All NMR Spectra were taken on either a Varion 56/60 or HA-lOO spectrometers, at room temperature, and as 102-302 carbon tetrachloride solutions. Integrations were made both electronically and with a planimeter. Gas Chromatography. Gas chromatographic analyses were preformed on an Aerograph A—90—P by using a 1/4" x 20' column packed with 20% carbowax 204M on 60/80-AW/HMDS-chromasorb W. Unless otherwise noted, preparative scale gas chromatography was done on the same instrument by using either the above column or a 1/4" x 6' column packed with the same material. Deamination Procedure. All deaminations were carried out in the following manner. A 0.100 mole of amine was dissolved in 75 ml of distilled water. The solution was contained in a round-bottomed flask, immersed in an ice bath, and fitted with an addition funnel, condenser, nitrogen inlet tube, and magnetic stirrer. A mixture of 0.130 moles of perchloric acid (as standard 70% solution) and 25 m1 H20 was added, and the resulting salt solution stirred at 0° for one-half hour. Deamination was then accomplished by very slow addition of 0.210 moles of sodium nitrite in 32 ml of water. The reaction was warmed to room temperature and stirred for four - six hours. The entire procedure was carried out under a nitrogen atmosphere. 59 60 The reaction mixture was saturated with sodium chloride and extracted with S x 75 ml of ether. The ether solution was washed with a saturated sodium bicarbonate-sodium chloride solution, dried over sodium sulfate and concentrated to approximately 15 ml by careful distillation. This crude product was than analyzed and various fractions purified and isolated as needed by gas chromatography. Yields were normally between 70% and 80% with typical product composition as follows: * Z Unidentified Amine Z Alcohols Z Alkenes Components Presumably composed of nitrite esters and unreacted amine. Preparation gf_Trimethylsilyl Ethers. The trimethylsilyl ethers of alcohols to be analyzed by mass spectroscopy were prepared in the following manner. Approximately 100 micro-liters of alcohol was placed in a small tube equipped with a cold finger-reflux condenser, and immersed in an oil bath at 60° - 90°. To this was added 130 micro-liters of hexamethyldisilazane followed by 10 micro-liters of dimethylchlorosilane,.and the mixture was allowed to react for twelve to twenty-four hours. The resulting trimethylsilyl ether 61 was purified by preparative gas chromatography. Preparation 2£_Cyclopentyl Cyanide. A mixture of 15.77 g (0.32 moles) of sodium cyanide and 95 ml of dimethylsulfoxide (freshly distilled from calcium hydride) was heated to 70° - 80° in a 500 m1 3-necked flask, fitted with a mechanical stirrer, reflux condenser and addition funnel. To this was added 40.00 g (0.27 moles) of cyclopentyl bromide over a period of three hours and the resulting mixture was stirred at 70° - 80° for an additional three hours. After the reaction mixture was cooled and a vacuum distillation head attached, the product was distilled under water aspirator pressure. Approximately 60 m1 of material was collected between 63° and 76°. The resulting nitrile was purified on an F & M Model 760 prepara- tive gas chromatograph by using a l" x 10" carbowax 20-M column at 140°. This resulted in 14.30 g (55% yield) of pure cyclopentyl cyanide. Preparation gf_Cyc1opentylcarbiny1 Amine. To 2.5 g (0.066 moles) of lithium aluminum hydride in 30 m1 of ether 6.0 g (0.063 moles) of cyclopentyl cyanide dissolved in 20 ml of ether was added over a period of forty minutes. The reaction was contained in an ice-cooled, 3-necked flask fitted with an addition funnel, stirrer, condenser, and drying tube. The mixture was stirred for three hours at room temperature, cooled in ice, and hydrolyzed by dr0pwise addition of 4 ml water fol- lowed by 4 ml of 52 sodium hydroxide solution. The ether layer was separated, dried over sodium sulfate, and carefully distilled. About 62 4.9 g (77% yield) of cyclopentylcarbinyl amine was collected at 135° - 136°. Preparation g£_Trans-24Methy1cyclopentanol. To an ice-cooled solution of 8.00 g (0.098 moles) of l~methylcyclopentene and 1.11 g (0.029 moles) of sodium borohydride dissolved in 50 m1 of tetrahydrofuran, 5.62 g (0.039 moles) of boron trifluoride etherate dissolved in 10 ml of tetrahydrofuran was added over a period of thirty minutes. The resulting mixture was warmed to room temperature and stirred for three hours. The entire reaction was carried out under a nitrogen atmosphere. Next, the reaction vessel was cooled in ice and 2 ml of water was added, followed by 10.0 ml of 5% sodium hydroxide and 10.0 mi of 30% hydrogen peroxide. The reaction mixture was stirred for several hours, the THF layer was removed, dried over magnesium sulfate, and evaporated. Distillation of the residue at water aspirator pressure yielded 3. 18 g (32% yield) of traps:2-methy1cyclopentanol, collected at 65° - 70°. Preparation gf_N-(l-methylcyclopentyl) Formamide (54). To 38 ml of glacial acetic acid contained in a 500 ml round-bottomed flask fitted with a stirrer, condenser, addition funnel, and thermometer was added 28.5 g of 1-methy1cyclopentene (0.303 moles) and 5.3 g of sodium cyanide (0.303 moles). Next, 78.6 g of sulfuric acid dissolved in 38 ml of glacial acetic acid was added over one hour while the reaction temperature was maintained at 40° - 50° by use of ice bath cooling. 63 After the resulting mixture was stirred for eighteen hours at room temperature, it was then diluted with 300 m1 of water, neutralized with a NaOH/HZO solution, and extracted with 5 x 100 m1 of ether. The ether solution was dried over magnesium sulfate, the ether was evaporated, and the residue was distilled at 12 - 15 mm pressure to give 27.7 g (63% yield) of N-(ldmethylcyclopentyl) formamide, boiling at 130° - 135°. Preparation 2f ldmethylcyclopentyl Amine. A solution of 31.78 g (0.250 moles) of N-(ldmethylcyclopentyl) formamide and 118 g potassium hydroxide in 480 m1 of water was refluxed for six hours. The reaction mixture was then steam distilled, and the distillate was salted and extracted with 4 x 100 m1 of ether. The ether solution was dried over sodium sulfate and after distillation yielded 10.34 g (42% yield) of l-methylcyclopentyl amine, collected at 112° - 114°. Preparation of Cyclopengylaminomethane-1,l-d An amount of 2.00 g 2. (0.0477 moles) of lithium aluminum deuteride was mixed with 30 m1 of ether (freshly distilled from lithium aluminum hydride) contained in a 3-necked round-bottomed flask, fitted with a stirrer, addition fun- nel, condenser, and drying tubes, and immersed in an ice bath. To this was added, over an hour, 4.60 g (0.0485 moles) of cyclopentyl cyanide dissolved in 20 ml of ether (freshly distilled from lithium aluminum hydride). The resulting mixture was stirred at room temper- ature for four hours. The reaction was again cooled in an ice bath and 4 m1 of water 64 followed by 4 ml of a 52 sodium hydroxide solution was added. Stir- ring was continued for 12 hours. The ether layer was then removed, dried over sodium sulfate, stripped, and distilled to yield 3.41 g (71% yield) of cyclopentylaminomethane-l,1-_c1_2 boiling at 136° - 138°. Preparation g£_Cyclopentanecarboxylic Acid. A hot (65° - 70°) solu- tion of 63.2 g (0.40 moles) potassium permanganate in 330 ml water was added over two hours to 30.0 g (0.30 moles) of cyclopentylcarbinol dissolved in 100 m1 of water contained in a 3-necked round-bottomed flask fitted with a stirrer, condenser, thermometer, and heated addi- tion funnel. The reaction was stirred for an additional two hours at 50° - 60°. The mixture was filtered and concentrated to approximately 150 m1 on a rotary evaporator and the resulting solution was acidified and extracted with ether. The ether solution was dried over magnesium sulfate and distilled at 12 - 15 mm pressure. The fraction boiling between 95° - 113° was collected and redistilled to give 14.0 g (33% yield) of cyclopentanecarboxylic acid, collected at 112° - 113°. Preparation of Cyclopentylmethanol-l,l-d A solution of 6.5 g 2. (0.058 moles) of cyclopentylcarboxylic acid dissolved in 25 ml of ether (freshly distilled from lithium aluminum hydride) was added over a period of forty minutes to a mixture of 2.4 g (0.058 moles) lithium aluminum deuteride and 40 m1 of ether (freshly distilled from lithium aluminum hydride). The reaction was stirred for sixteen hours and hydrolyzed by slow 65 addition of S‘ml water followed by 5‘ml of a 5% sodium hydroxide solution. After an additional three hours stirring, the ether layer was removed, dried over magnesium sulfate and distilled at 12 - 15 mm to yield 4.64 g (82% yield) of the desired alcohol, collected at 80° - 85° 0 Preparation g§_Diborane (55). A solution of 47.5 g sodium borohydride in 900 ml of diglyme (freshly distilled from lithium aluminum hydride) was added over a period of four to five hours to 337 g of boron tri- fluoride etherate (freshly distilled from calcium hydride). The resulting diborane was bubbled through a solution of sodium borohydride in diglyme to remove traces of boron trifluoride and into 700 ml of ice-cooled tetrahydrofuran. After the addition of sodium borohydride was complete, the generating flask.was heated to 50° for an additional two hours. The entire reaction was carried out under nitrogen using a mercury safety valve, and a mercury/acetone trap to remove any es- caping diborane, as described in Reference 55. The resulting BZH6/THF solution was found to be 1+0.l molar by titration with a tetrahydro- furan/water solution. Preparation gf_A3-cyclopentenol (56). A one molar solution of diborane in tetrahydrofuran (400 ml, 0.8 moles of BH3) was added over a period of fifteen minutes to an ice-cooled 1 liter flask containing 239.5 g of s-pinene (1.76 moles, 10% excess, freshly distilled from lithium aluminum hydride). The mixture was stirred at 0° for two hours and 105.7 g (1.6 moles) of freshly distilled cyclopentadiene was added and 66 stirring continued at room temperature for thirty hours. The entire reaction was carried out under nitrogen. The reaction vessel was cooled in ice and 30 ml of water added to decompose any excess hydride. The organoborane was oxidized by dr0pwise addition of 256 m1 of sodium hydroxide and 256 m1 of hydrogen peroxide. The aqueous layer was well salted, and the organic layer removed. Excess cyclopentadiene and tetrahydrofuran were removed under vacuum, and the resulting mixture diluted with 200 m1 of ether. This solution was then stirred for one hour with 750 ml of 1 molar aqueous silver nitrate. The organic layer was removed and extracted twice more with 200 m1 of silver nitrate solution. The water solutions were combined and washed twice with ether. An excess of sodium chloride was added to completely precipitate silver chloride, and the unsaturated alcohol extracted with ether. The ether solution was dried over magnesium sulfate, the ether removed, and the resulting material distilled to give 14.54 g (21.52 yield) of A3-cyclopentenol. Preparation 2£_Tris(triphepylphosphine)chloroRhodium (I) (51, 52, 53). Freshly recrystallized triphenylphosphine (6 g) dissolved in 130 ml of degassed ethanol was heated to reflux in a nitrogen atmosphere. Rhodium chloride trihydride (l g) was added, and refluxing continued for one-half hour. The mixture was filtered and the precipitate washed with degassed ether and dried under vacuum to yield 3.44 g (99% yield) of tris(triphenylphosphine)chloroRhodium (I). Preparation 9_f_Cyclopentanol-3,4-d2 (51, 52, 53). Rhodium chloride 67 triphenylphosphine (1.2 g) was placed in a 500 m1 filter flask fitted with a rubber septum cap. The flask was repeatedly evacuated and flushed with deuterium gas. Degassed benzene (375 ml) was injected and the solution stirred for one hour. A3-cyclopentenol (12.0 g) was injected and the reaction mixture stirred for five hours. The solvent was removed by distillation at atmospheric pressure, and then at reduced pressure to yield 9.29 g (71%) of cyclopentanol-3,4fid2. Preparation pfCyclopentyltosylate-3,4-d2 (57). p-Toluenesulfonyl chloride (121 g, 0.635 moles) was dissolved in 300 m1 of cold pyridine, and 28.29 g (0.318 moles) of cyclopentanol-3,4-lgl__2 added and the re- sulting solution allowed to react at 0° for 24 hours. The reaction mixture was then poured into 1500 ml of a stirred ice/water mixture and stirring continued for one-half hours. The tosylate was extracted five times with 300 m1 of ether, the ether solution washed with cold 1:1 hydrochloric acid/water, then with water, and dried over Na2804/K2C0 3. The ether was removed to yield 61.18 g (73.52) of crude tosylate. Preparation 2£_Cyclopenty1-3,4-d Cyanide. A solution of powdered 2 sodium cyanide (15.45 g, 0.315 moles) in 100 ml dimethyl sulfoxide (freshly distilled from CaHz) was heated to 50° - 60° and 61.18 g (0.252 moles) of crude cyclopentyltosylate added over a period of two hours. An additional portion of 50 m1 of dimethyl sulfoxide was added, and the reaction maintained at 50° - 60° for five hours. The resulting mixture was cooled to approximately 40° and set up for 68 distillation under water aspirator pressure. Distillation was con- tinued until approximately 60 ml had been collected. This material was redistilled twice more to yield 11.51 g (47.52 yield) of cyclo- penty1-3,4fg 2 cyanide. Preparation of l-aminomethylcyclopentane-3,4—d To an ice-cooled 2. mixture of 5.90 g (0.155 moles) of lithium aluminum hydride and 70 m1 of ether, 11.52 g (0.119 moles) of cyclopentyl-4,4122 cyanide in 50 ml of ether was added over a period of approximately one hour. The reaction was stirred at room temperature for eighteen hours. cooled in ice, and carefully hydrolyzed with 14 m1 of water followed by 14 m1 of a 52 sodium hydroxide solution. The resulting mixture was stirred at room temperature for six hours. The ether layer removed, dried over sodium sulfate, and dis- tilled to yield 7.32 g (61% yield) of 1-aminomethy1cyclopentane-3,4fd2. Preparation gf_Cyclopentanecarboxylic Acid-3,4-d2. A solution of 3.32 g (0.0342 moles) of cyclopentyl-3,4fid2 cyanide, 3.74 3 (0.0935 moles) of sodium hydroxide and 40 m1 of a 70% ethanol - 302 water solution was refluxed for four hours. Ethanol was removed on a rotary evaporator and the residue acidified with a 502 hydrochloric acid solution. The reaction mixture was extracted with 3 x 30 ml ether, the ether solution dried over magnesium sulfate, and evaporated to yield 2.70 g (68% yield) of the deuterated acid. 69 Preparation 9: Cyclopentylcarbinol-3,4-d A solution of 2.70 g 2. (0.0233 moles) of cyclopentanecarboxylic acid-3,4ed2and 10 ml of ether was slowly added to an ice-cooled mixture of 1.50 g (100% excess) lithium aluminum hydride and 50 m1 of ether. The reaction was stirred at room temperature for fifteen hours and hydrolyzed by care- ful addition of 5 ml of water followed by 5 ml of a 52 sodium hydroxide solution. The ether layer was removed, dried over magnesium sulfate, and evaporated to yield 1.86 G (91% yield) cyclopentylcarbinol-3,4fid2. Preparation g£_§yclopentanol-1-d. A solution of 23.90 g (0.286 moles) of freshly distilled cyclopentanone and 50 m1 ether (freshly distilled from lithium aluminum hydride) was added over a period of ninety minutes to an ice-cooled mdxture of 3.00 g (0.0715 moles) lithium aluminum deuteride and 100 m1 of ether (freshly distilled from lithium aluminum hydride). The reaction was stirred at room temperature for sixteen hours and hydrolyzed by careful addition of 6 ml water followed by 6 m1 of a 5% sodium hydroxide solution. The ether layer was separated, dried over sodium sulfate, and evaporated to yield 22.97 g (92%) of deuterated alcohol. Preparation g£_cyclopentyltosylate-l-d. An ice cold solution of 100.00 g (0.528 moles) of p-toluenesulfonylchloride in 320 ml of pyridine was combined with 22.97 g (0.264 moles) of cyclopentanol-lfig and the reaction allowed to stand at 0° for twenty-four hours. The mixture was then poured into 2000 ml of ice water and stirred 70 vigorously. The resulting crystals were collected by vacuum filtra- tion, washed with water, and dried in a vacuum desicator to yield 38.9 g (61% yield) of cyclopentyltosylate-led, Preparation gf chlopentyl-l-d Cyanide. A mixture of 38.9 g (0.161 moles) of cyclopentyltosylate-lfd_, 9.46 g (0.193 moles) of powdered sodium cyanide, and 200 m1 of dimethylsulfoxide (freshly distilled from calcium hydride) contained in a 500 m1 round-bottomed flask, fitted with a stirrer and reflux condenser, was allowed to react at room temperature for eighteen hours, and then at 50° - 60° for four hours. This material was distilled at 12 - 15 mm until approximately 110 m1 of distillate had been collected. The impure cyanide was re- distilled three additional times resulting in 12.11 g of 702 pure (by VPC) cyclopentyl-1fg_cyanide. Boiling range: 54° - 62° at 12 - 15 mm. Yield - 55%. Preparation of l-aminomethy1cyclopentane-1-d. To a mixture of 7.18 g (0.189 moles) of lithium aluminum hydride in 250 m1 of ether, 12.11 g of 70% pure cyclopentyl-lfd_cyanide (0.126 moles) dissolved in 30 ml of ether was added over a period of thirty minutes at 0° and the reaction stirred at room temperature for eighteen hours. Hydrolysis was accomplished by careful addition of 16 ml water followed by 16 m1 of a 5% sodium hydroxide solution. The ether layer was removed, dried over sodium sulfate, and distilled to give 7.91 g (91% yield) of 1-aminomethylcyclopentane-led 71 collected at 135° - 139°. Preparation 2£DCyclohexanol-1-d. A solution of 6.85 g (0.070 moles) of freshly distilled cyclohexanone and 20 m1 of ether (freshly dis- tilled from lithium aluminum hydride) was carefully added to an ice-cooled mixture of 1.00 g (0.0238 moles) lithium aluminum deuteride and 30 ml of ether (freshly distilled from lithium aluminum hydride). The reaction was stirred at room temperature for eighteen hours and then hydrolyzed by careful addition of 2 ml water followed by 2 m1 of a 52 sodium hydroxide solution. The ether layer was separated, dried over magnesium sulfate, and evaporated to yield 6.1 g (87% yield) of cyclohexanol-l-d, The crude alcohol was purified by preparative gas chromatography as needed. 10. ll. 12. 13. 14. REFERENCES C. L. Collins, Chem. Rev., 543 (1969). J. D. Roberts and M. Halmann, 8, 52, Chem. Soc., 18, 759 (1953). P. S. Skell and I. Starer, 8, Am. Chem. Soc., 81, 4117 (1959). (b) P. S. Skell and I. Starer, 8, gm, Chem. Soc., 81, 2971, (1960). (c) P. S. Skell and I. Starer, 8, Am, Chem. Soc., 83, 3962, (1962). 0. A. Reutov and T. N. Shatkina, Tetrahedron, 18, 237 (1962). O. A. Reutov and T. N. Shatkina, Izv. Akad. Nauk. Sssr. Otd. Khim, Nauk., 195 (1962). (Bull. Acad. U.S.S.R. Div. Chem. Sci. [Eng.] 180 (1962). G. J. Karabatsos and C. E. Orzech, Jr., 8, 59. Chem. Soc., 84, 2838 (1962). (a) R. L. Baird and A. A. Aboderin, Tet. Letters, 234 (1963). (b) R. L. Baird and A. A. Aboderin, 8, gm, Chem. Soc., 88, 252 (1964). R. L. Baird and A. A. Aboderin, 8, 8m, Chem. Soc., 86, 2300 (1964). C. C. Lee and J. E. Kruger, 8, Am, Chem. Soc., 87, 3986 (1965). C. C. Lee, J. E. Kruger and E. W. C. WOng, 8, Am, Chem. Soc., 81, 3985 (1965). G. J. Karabatsos, C. E. Orzech, Jr. and S. Meyerson, 8, 52, Chem. Soc., 81, 4394 (1965). C. C. Lee, J. E. Kruger, Tetrahedron, 18, 2539 (1967). C. C. Lee, Kwok-Ming Wan. 8, 82, Chem. Soc., 31, 6416 (1969). G J. Karabatsos, C. E. Orzech, Jr., J. L. Fry, S. Meyerson, 8, Am, Chem. Soc., 21, 606 (1970). 72 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 73 L. Friedman, and A. Jurewicz, 8, 82, Chem. Soc., 91, 1800 (1969). G. J. Karabatsos, J. L. Fry, 8. Meyerson, Tet. Letters, 3735, (1967). G. J. Karabatsos, J. L. Fry. S. Meyerson, 8, 82, Chem. Soc., 81, 614 (1970). C. C. Lee, D. J. Whodcock, 8, 82, Chem. Soc., 81, 5992 (1970). G. J. Karabatsos, C. Zioudrou, S. Meyerson, 8, 82, Chem. Soc., 81, 5996 (1970). R. Hoffman, 8, Chem. Phys., 88, 2480 (1964). J. D. Petke and J. L. Whitten, 8, 82, Chem. Soc., 88, 3338 (1968). H. Fischer, H. 0. Smith, K. Miller, and H. Kollmar, Tet. Letters, 5821 (1968). H. Kollmar, and H. 0. Smith, Tet. Letters, 1833 (1970). Nicolae Bodor and M. J. S. Dewar, 8, 82, Chem. Soc., 88, 6685 (1971). L. Random, V. Buss, J. A. Pople, P. v. R. Schleyer, 8, 82, Chem. Soc., 88, 1813 (1971). G. J. Karabatsos, Nelson H31 and Seymour Meyerson, 8, 82, Chem. Soc., 88, 5649 (1966). G. J. Karabatsos, Raymond A. Mount, Donald 0. Rickter, Seymour Meyerson, 8, 82, Chem. Soc., 88, 5651 (1966). Ibid. 8, 82, Chem. Soc., 81, 1248 (1970). J. H. Bayless, F. D. Mendicino, L. Friedman, 8, 82, Chem. Soc., 81, 5790 (1965). A. T. Jurewicz and L. Friedman, 8, 82, Chem. Soc., 88, 149 (1967). G. J. Karabatsos, Nelson H51 and Seymour Meyerson, 8, 82, Chem. Soc., 81, 621 (1970). Lester Friedman, A. T. Jurewicz, and J. H. Bayless, 8, 82, Chem. Soc., 91, 1795 (1969). Lester Friedman and A. T. Jurewicz, 8, 82, Chem. Soc., 81, 1803 (1969). P. S. Skell and R. J. Maxwell, 8, 82, Chem. Soc., 84, 3963 (1962). 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 74 G. J. Karabatsos, Mona Anand, D. 0. Rickter, S. Meyerson, 8, 82, Chem. Soc., 81, 1254 (1970). 0. A. Reutov, Yu G. Bundel, 1. Yu. Levona, I. R. Prokhoronko, Dokl. Akad. Nauk. Sssr., 188, 348 (1969), (Eng. 732). G. J. Karabatsos, C. Orzech, Jr., S. Meyerson, 8, 82, Chem. Soc., .88, 1994 (1964). M. S. Silver, 8, 82, Chem. Soc., 88, 3482 (1961). P. S. Skell, I. Starer, A. P. Krapcho, 8, 82, Chem. Soc.,_81, 5257 (1960). M. S. Silver, 8, Org. Chem., 18, 1689 (1962). M. S. Silver, J. 82, Chem. Soc., 82, 2971 (1960). G. J. Karabatsos, L. J. D. Graham, 8, 82, Chem. Soc., 81, 5250 (1960). N. C. Deno, D. LaVietes, J. Mockus, P. C. Scholl, 8, 82. Chem. Soc., 88, 6457 (1968). M. Saunders, E. Hagen, 8, 82, Chem. Soc., 90, 6881 (1968). M. Saunders, E. Hagen. J. Rosenfeld, 8, 82, Chem. Soc., 88, 6882 (1968). M. Saunders, J. Rosenfeld, 8, 82, Chem. Soc., 81, 7756 (1969). G. M. Kramer, 8, 82, Chem. Soc., 91, 4819 (1969). (a) T. N. Shatkina, E. V. Leontieva, 0. A. Reutov, Dokl. Akad. Nauk. Sssr., 177, 143-4 (Eng. 1006) (1967). (b) Ibid. 111, 373-5 (Eng. 1047) (1967). (c) 0. A. Reutov, Yu G. Bundel, V. A. Savin, M. N. Ryabtsev, Dokl. Akad. Nauk. Sssr., 165, 1303 (Eng. 1198) (1965). (d) 0. A. Reutov and T. N. Shatkina, Dokl. Akad. Nauk. Sssr., 142, 835, (Eng. 98) (1962). (e) A. A. Lubovich, Yu G. Bundel, U. A. Saviv and 0. A. Reutov, Dokl. Akad. Nauk. Sssr., 165, 1078 (Eng. 1180) (1965). (a) 0. E. Edwards and M. Lesage, Can 8, Chem., 41, 1592 (1963). (b) 0. E. Edwards and M. Lesage, Chemistry and Industry, 1107 (1960). 75 50. (a) T. N. Shatkina, E. V. Leontieva, O. A. Reutov, Dokl. Akad. Nauk. Sssr., 173, 113 (Eng. 224) (1967). (b) Yu G. Bundel, K. G. Dankratuva, N. I. Shutilova, and O. A. Reutov, Dokl. Akad. Nauk. Sssr., 194, 1323 (Eng. 752) (1970). 51. A. J. Birch and K. A. M. Walker, 8, Chem. Soc., (c), 1894 (1966). 52. J. F. Young, J. A. Osborn, F. H. Jardine, and C. Wilkinson, Chemical Comm., 131 (1965). 53. J. A. Osborn, F. H. Jardine, J. F. Yound and C. Wilkinson, 8, Chem. Soc., (A), 1711 (1966). 54. J. J. Ritter and P. D. Minieri, 8, 82, Chem. Soc., 18, 4045 (1948). 55. "Hydration of Olefins and Acetylenes via Hydroboration," by Beorge Zweifel and H. C. Brown in Organic Reactions, Vol. 13, pp. 31-33. 56. Henry M. Hess and H. C. Brown, 8, Org. Chem., 81, 4138 (1967). 57. Reagents for Organic Synthesis, by L. F. Fieser and M. Fieser, pp. 1180-81. ”'7ITI'I‘flfil7ujjfl‘jflfialjfiflallfy'ffliulfijlflifllflfllfl“