THE DEAz‘fiIHATIQN U‘F EEOWPICALLY-LABELED BUTYLAMIfiES Thesis for the Degree of Ph. D. WOMAN STATE UNIVERSH‘.’ RAMON A. MOUNT 1967 ’4 6.815 This is to certify that the thesis entitled THE DEAMINATION OF ISOTOPICALLY-LABELED BUTYLAMINES presented by Ramon A. Mount: has been accepted towards fulfillment of the requirements for PhoD 0 degree in ChEIlliStry Mafia Major professor Date August 31, 1967 0-169. LIBRARY MlChigan 3C3” ‘ University ABSTRACT THE DEAMINATION OF ISOTOPICALLY-LABELED BUTYLAMINES by Ramon A. Mount The intermediacy of protonated cyclopropanes in the deamination of l-propylamine has been ascertained by iso- tope-position rearrangement of the carbon skeleton and by formation of cyclopropane. Such intermediates have not been detected in the deamination of neopentylamine. Aliphatic amines which are structurally intermediate between l-propylamine and neopentylamine undergo rearrange- ments that have not been fully investigated. Thus, although a small amount of methylcyclopropane formation accompanies the deamination of 1-butylamine, 2-butylamine and isobutyl- amine, isotope-position rearrangement in these and similar systems is less documented. Accordingly, the present in- vestigation was undertaken in order to assess the effects on deamination reactions of methyl substitution at C-1 and C-3 of the base 1—propyl model. To this end, isotopically- labeled 1- and Z-butylamines were prepared and deaminated under standard conditions. Analysis of product alcohols served as the basis for conclusions derived from this study. Three labeled l-butylamines, dideuterated at the 1-, 2- and 3-positions, respectively, gave l—butanol and 2-butanol as the only alcoholic products. Mass spectral data showed that, in each case, the isotopic distribution of the 1- butanol formed was identical to that of the starting amine. This finding permits mechanistic pathways involving Ramon A. Mount 1,4-hydride shifts, 1,2-ethyl shifts and 1,3-methyl shifts to be excluded. Also, symmetrical bridged ions, such as i, cannot be important in the formation of 1-butanol. Equilibration of protonated cyclopropanes such as g and g, $H3 ?H3 ?H3 /Cfxi2 CH\- > CEI\‘ / + \ / \\ +:H <___.. / \ +‘/~D / \ \ I \ I CH2 ---CD2 CH2-——CD2 CH2‘-——CHD 1, Z, 3 with subsequent formation of 1-butanol, is likewise dis- allowed, as is any mechanism involving interconverting primary and secondary cations. In each of the deaminations of the labeled l-butylamines, about 75-81% of the 2-butanol formed was the result of a 1,2-hydride (or deuteride) shift followed by solvent cap— ture of the resulting 2-butyl cation. Establishment of the mechanistic pathways(s) by which the remaining 2-butanol was produced was of interest, because of the possibility of a 1,3-hydride (or deuteride) shift occurring gig carbon- or hydrogen-bridged intermediates (§;g,, £72). Mass spectral analysis of the 1—butyl—3,3j§2 system (eq. 1), coupled with nmr data, permitted exclusion of mechanisms involving a nominal 1,3—hydrogen migration. By establishing that all of the product 2—butanols had deuteria only at C-2 and C-3, alcohols (fi'and Z) arising from a 2-butyl cation formed by an initial 1,3-deuteride shift was ruled out. All of the detectable Z-butanol formed by deamination of Ramon A. Mount CH3CD2CHOHCH3 CH3CDOHCHDCH3 i E f f 1'2'H“3 CHacndeCH3 E'Z‘DTL cHadDCHDCH3 CH3CD2CH2CH2NH2 —>‘ 1,3- ~' + «———JQ—> CH3CDCH2CH2D _,~ + L—g—Ii» CH 3CHDCHCH2b < 1 l CH3CDOHCH2CH2D CH3CHDCHOHCH2D 6 7 N N the labeled l-butylamines was the result of one or more 1,2-hydride (or deuteride) shifts. The deaminations of both 2-butyl-2-gfamine and 2-butyl- 3,3-g2-amine gave 2-butanol as the only alcohol. In these systems, also, only 1,2-hydrogen migrations were observed. Again, there were no 1,3—shifts and no interconverting primary and secondary cations. No evidence for bridged-ion intermediates was obtained. It was also established that in the deamination of 1- and 2—butylamine (a) alcohol formation is irreversible; (b) alcohols are formed from unrearranged or rearranged carbonium ions that have precursors derived from the starting amines, not from other reaction products such as olefins; and (c) mechanisms involving diazoalkane or carbene forma— tion are not significant. THE DEAMINATION OF ISOTOPICALLY‘LABELED BUTYLAMINES BY Ramon Ax Mount A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1967 To Anne P. P. ACKNOWLEDGMENTS The author extends his sincere appreciation to Professor G. J. Karabatsos for his guidance and encourage- ment during the course of this investigation. The author is indebted to Mr. Seymour Meyerson of the American Oil Company, Whiting, Indiana, for performing mass spectral analyses. Financial support from the Dow Chemical Company, the National Science Foundation and the Petroleum Research Fund is gratefully acknowledged. Finally, the author wishes to thank his parents, who taught him the value of education. iii TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . I. General . . . . . . . . . . II. Mass Spectral Data . . . III. 1-Butyl Systems . . . . . . A. 1-Butanol Formation . B. 2-Butanol Formation . IV. 2—Butyl Systems . . . . . . V. Summary . . . . . . . . . . EXPERIMENTAL . . . . . ... . . . . . . I. General . . . . . . . . . . II. Preparation of 1-Butyl-1,l-§2-ammonium Perchlorate . . . . . . III. Preparation of 1-Butyl-2,27§2-ammonium Perchlorate . . . . . IV. Preparation of 1-Butyl—3,3-g2-ammonium Perchlorate . . . . . . V. Preparation of 2-Buty1-2-grammonium Perchlorate . . . . . . VI. Preparation of 1—Butyl-3,3-§2-ammonium Perchlorate . . . . . .VII. Preparation of Authentic, Deuterated Alcohols . . . . . . . . VIII. Deamination of Butylamines REFERENCES . . . . . . . . . . . . . . iv Page 16 16 19 38 42 44 51 74 76 76 76 77 79 82 82 85 87 89 LIST OF TABLES TABLE Page I. Mass Spectrum of CH3CH2CH2CHZOSi(CH3)3 . . 20 II. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)3 . 21 III. Mass spectrum of authentic CH3CH2CH2CD(CH3)OSi(CH3)3 . . . . . . . . 22 Iv. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 2—butanol recovered from blank deamin- ation with 2-butanol-2-g_ . . . . . . . . 23 V. Mass Spectrum of CH3CH2CH2CHZOSi(CH3)3 from 1-butanol obtained from the deamination of 1-butyl-1,1—g2-ammonium perchlorate . . . 25 VI. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 2-butanol obtained from the deamination of 1-butyl-1,1—d3-ammonium perchlorate . 26 VII. Mass spectrum of authentic CH3CH2CH2CDZOSi(CH3)3 . . . . . . . . . . 27 VIII. Mass spectrum of authentic CH3CH2CH(CHD2)OSi(CH3)3 . . . . . . . . . 28 IX. Mass Spectrum of CH3CH2CH2CHZOSi(CH3)3 from l-butanol obtained from the deamination of l-butyl-2,2—d2-ammonium perchlorate . . . 29 X. Mass spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 2-butanol obtained from the deamination of 1-butyl-2,2—g2-ammonium perchlorate . 30 XI. Mass Spectrum of authentic CH3CH2CD2CHZOSI(CH3)3 . . . . . . . . . . 31 XII. Mass spectrum of CH3CH2CH2CHZOSi(CH3)3 from 1-butanol obtained from the deamination of 1-butyl—3,3-g2-ammonium perchlorate . . . 32 XIII. Mass spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 2-butanol Obtained from the deamination of 1-butyl-3,3-§2—ammonium perchlorate . 33 XIV. Mass spectrum of authentic CH3CD2CH2CHZOSi(CH3)3 . . . . . . . . . . 34 LIST OF TABLES (Cont.) TABLE XV. XVI. XVII. XVIII. XIX. XXII. IXXIII. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 2-butanol obtained from the deamina- tion of 2-butyl-25gfammonium perchlorate . Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 2-butanol obtained from the deamina- tion of 2-butyl-3,3-g3-ammonium perchlorate Mass Spectrum of authentic CH3CD2CH(CH3)OSi(CH3)3 .. . . . . . . . . . Results of mass spectral analyses of the trimethylsilyl ethers of authentic, deuter- ated alcohols . . . . . . . . . . . . . . . Results of mass Spectral analyses of the trimethylsilyl ethers of deamination product alcohols . . . . . . . . . . . . . Integrated nmr signals of methyl protons of the trimethylsilyl ethers of 2-butanol ob- tained from the deamination of 1-butyl-3,3- gz-ammonium perchlorate and of unlabeled 2-butanol.. . . . . . . . . . . . . . . . . Integrated nmr signals of various protons of the 2-butanol obtained from the deamination of 1—butyl-3,3-g2-ammonium perchlorate . . Integrated nmr Signals of various protons of the 2—butanol obtained from the deamination of 2-buty1-2-gfammonium perchlorate and of unlabeled 2-butanol . . . . . . . . . . Integrated nmr Signals of various protons of the 2-butanol obtained from the de— amination of 2-butyl-3,3-§2-ammonium per- chlorate and of unlabeled 2-butanol . . vi Page 35 36 37 39 4O 49 49 55 58 LIST OF FIGURES FIGURE Page 1. Synthesis of 1-butyl-2,2-§2-amine . . . . . . 17 2 Synthesis of 1-butyl-3,3-g2-amine . . . . . . 17 3. Synthesis of 2-butyl-3,3-§2—amine . . . . . . 18 4 Possible rearrangement pathways of the 1-butyl cation . . . . . . . . . . . . . . . 41 5. Methyl region of the nmr spectrum of the tri- methylsilyl ether of the 2-butanol obtained from 1—butyl-3,3-§2—amine . . . . . . . . . . 48 6. Nmr Spectrum of the 2-butanol obtained from 1-butyl-3,3-g3-amine . . . . . . . . . . . . . 51 7. Possible rearrangement pathways of the 2-butyl cation . . . . . . . . . . . . . . . . . . . 52 8. Methylene and methyl regions of the nmr Spec- trum of the 2-butanol obtained from 2-butyl- Z-Qfamine . . . . . . . . . . . . . . . . . . 54 9. 1-Methyl region of the nmr Spectrum of the trimethylsilyl ether of the 2-butanol ob- tained from 1-butyl-2—gfamine . . . . . . . . 57 10. Methyl region of the nmr Spectrum of the 2- butanol obtained from 2-butyl-3,3-§2-amine . 6O 11. Proposed mechanisms of formation of the pro- ducts of the deamination of 1-butylamine . . 68 12. Ratios of labeled 2-butanols obtained from the deaminations of deuterated 1-butylamine, 2-butylamine and isobutylamine . . . . . . . 71 vii INTRODUCTION Throughout the history of organic chemistry new and novel hypotheses have been advanced intended to help chemists rationalize and understand their experimental data. These new ideas often could be reported only by using new struc— tural notation which frequently was the subject of test, modification, expansion and controversey to an extent equal to that of the theory being described. An example of such an idea, which has been in the pro- cess of development for thirty years, is shown by the bridged cationic Species 1; Such notation was used as early as 1937 . . -C-—C- \+/ X 1 by Roberts and Kimball (1) to describe a bromine-ethylene, or bromonium, complex ion (1” X = Br). In 1938 Winstein and Lucas (2) used l/ X = Ag, to —>C/ +1} —> CH314CH2CH20H + CH3CH214CH20H H§:_ CH2 3 the label at C-2 in the rearranged 1-propanol, rather than at C-3, which would be the result of a 1,3—hydride Shift prior to capture of the 1-propyl cation by solvent. Compet- ing with this, of course, were mechanistic pathways leading to the formation of 2-propanol, propylene and unrearranged 1-propanol as well as nitrites and nitrates that are char— acteristically produced during deamination of aliphatic amines (7). Thus an aliphatic carbon-bridged ion was judged impor; tant in the propyl system, though not as Significant as an aronium bridge in the deamination of 2-phenylethylamine. The latter intermediate was reported (8,9) to be the pre- cursor of 54% of the 2—pheny1ethanol produced, as opposed to a value of 17% in the propyl system (5). The methyl-bridged ion 4, and the related protonated cyclopropane 5/ were given roles as short-lived intermedi- ates by Skell and Starer when they reported cyclopropane as a hydrocarbon product of the deamination of 1-propyl- amine (10). CH3 CH3 CH2 + / \ H > // +\\ < > X > CH2_—CH2 + CH2-=;CH2 Griz—CH2 g 5 "' 2 CH2 + ( ) / \ + H CH2—_CH2 The protonated cyclopropane Q'was suggested to be an intermediate in the formation of isotope-position rearranged 4 products as well as the cyclopropane, although these workers later explained (11) the reaction in terms of equilibrating 1-propyl cations. Formation of cyclopropane was accommodated by a 1,3- ring closure by way of the ion Q3 Results were also reported at that time which indicated that the carbenoid intermediate CH3CH2CH: was not involved in the "major pathway to cyclopropane". Reutov and Shatkina (12) repeated the experiment of Roberts and Halmann with 1-propyl-1-14C—amine. In contrast to the earlier work, in which it was concluded that the 14C label moved to C-2 in the rearranged 1-propanol, Reutov and Shatkina reported that the 14C label was exclusively at C-1 and C-3. Two possible explanations were advanced, neither of which included nonclassical intermediates. The first in- volved 1,3-hydride migration, perhaps by way of 6, CH2 > Q§2\;?CH2 CH3CH2 1 4CH2 + + _ > CH2CH214CH3 (3) \ + . \ I H 6 N The second consisted of two successive 1,2-hydride shifts, the intermediate being a 2-pr0py1 cation. 5 + .__i CH2CH214CH3 (4) < + CH3CH2 14CH2+ :—-> CH3CH14CH3 Sequence 4 was precluded by the results of Karabatsos and Orzech (13) concerning the deamination of 1-propyl- 1,1,2,2-§A-amine, as no Significant amount of protium was found to be incorporated at C-2. Their results agreed with a nominal 1,3-Shift mechanism. Reutov (14) also gave evidence for a 1,3-hydride shift mechanism with the deamination of 1-propyl-21E-amine. The 1—propanol product was reported to contain tritium only at C-2. In 1963 Baird and Aboderin (15) reported new evidence accommodating the existence of protonated cycloprOpaneS as true intermediates. Treatment of cyclopropane with 7.5 M deuteriosulfuric acid gave significant amounts of monodeuer— ated and dideuterated cyclopropane, along with solvolysis products. Although a chain mechanism embracing classical ionic structures could not be excluded, a mechanism in- volving reversible formation of cyclopropane-H+-v-complexes attractively rationalized the observations. CH2—-CH2 CHz—CI-Iz CH2-—CHD CH2 + H+ Results of a study of the solvolysis products were published in 1964 (16). The 1-propanol produced was found 6 to have an average deuterium distribution of 0.38, 0.17 and 0.46 deuterium atom in the 1-, 2-, and 3-positions, re- spectively. It was also observed that 1-propanol, 2-propanol and propylene were formed in relative amounts of 1:0.0027: 0.0005, in contrast to values of 1:4.6:4 obtained in the deamination of 1-propylamine (17). These data precluded a mechanism involving equilibrating primary carbonium ions as the major solvolysis pathway. On the other hand, they could be fitted to a mechanism composed of equilibrating carbon- and hydrogen-bridged cations, which was also consistent with the observed formation of 1—propy1 solvolysis products (sequence 5). CHZD ,C§3 I / +\\ I + \ l \ I \ CH2- - - CH2 CHD=CH2 \ + 9+ Jr \\ 7/ \\ (5) CH2 CH2 .\\3 CHZ“ c \ CH21~ /\\ +‘D / \\ +\H / \\ +31 I l/ \ I CH2—-CH2 CH2——><:H2 CED—CH2 50H; SOHl SOHl CHZDCH2CHZOS cnacnzcnnos CH3CHDCHZOS + CH2DCH2CHZOS While the above scheme did not correlate with the results of Reutov and Shatkina, it could be used to interpret the findings of Aboderin and Baird (18) regarding the deamina- tion of 1-propyl-3,3,3-g3-amine. The cyclopropane produced, isolated by gas chromatography and analyzed by mass Spectrometry, l' 7 was found to consist of 43% 92‘ and 57% g3—Species. The high concentration of cyclopropane-g3 could not be rationalized by using equilibrating primary carbonium ions without invoking an unprecedented high kH/kD ratio. However, the results were explained by using a reasonable kH/kD (2.7 to 3.0) in conjunction with a mechanism involving formation of El hydrogen—bridged ion from a 1—propyl cation (or its im— mediate precursor) followed by equilibration with isomeric carbon—bridged ions. Communications published in 1965 by Lee and coworkers and by Karabatsos, Orzech and Meyerson supported the con- clusions of Baird and Aboderin. Lee, Kruger and Wong (19) examined the 1-propyl solvolysis product of the deamination of 1-propyl—2,2-§2-amine and 1—propyl-1—tfamine. Regarding the former, nuclear magnetic resonance spectroscopy quali— tatively indicated a protium increase at C-2. Approximately 3% of the 1-propanol from the tritiated amine was found to have tritium at C—2 and C-3, with roughly equal amounts at each position. Based on their finding 3-4% isotopic rear- rangement from C-l to C-2 and C-3, again in equal amounts, in the deamination of 1-propyl-1-14C—amine, Lee and Kruger (20,21) concluded 4-6% of the l—propanol formed had pro- tonated cyclopropane precursor. Karabatsos, Orzech and Meyerson concurred from results of their study (22) of the deamination of two isotopically— labeled 1-propylamines, dideuterated at C-1 and C—2, re- spectively. Their observations were in accord with a mechanism 8 involving equilibrating protonated cyclopropanes, not neces- sarily arising from methyl-bridged Species. Two other recent reports are relevant to the present discussion. Hart and Schlosberg (23) obtained the chloro- ketones Z” 8’and 9” along with the butenone 12/ upon acetyla- tion of cyclopropane with acetyl chloride and aluminum chloride. Intermediacy of protonated cyclopropanes accounted for the observations, as shown in eq. 6. SH. CH3coc=CH2 12, t-H+ CH2 CH2“ I,»CH2 \\ +‘fi If + / CH3COC1 + —> \ ’ —<-;>_ ‘ ll CH2'—CH2 CHz—‘C‘ CH2_—-(':H COCH3 coca3 (6) Cl-l Cl-l CH3COCH2CH2CH2C1 CH3cocHCH2c1 + Z. CH3 CH3COCHC1CH2CH3 2 §, Deno and Lincoln (24) published similar observations concerning bromination cf cyclopropane. In this instance, the products of the reaction were three isomeric' di-- bromopropanes. As the propyl system is the Simplest in which isotope— position rearrangements can be observed, one may classify higher homologs as a 1-pr0pyl system substituted at carbons 1, 2 and/or 3. For example, methyl substitution at C-1, C-2 9 and C-3 yields the 2-butyl, isobutyl and 1-butyl systems, respectively. In an important paper published in 1957, Cram and Mc- Carty (25) reported the first experimental evidence directly implicating a bridged-alkyl Species in a butyl system. These workers investigated the methyl-shift products of the de- amination of EEEEEF and erythro—3-phenyl—2—butylamine in glacial acetic acid. NH2 l 1,2- ... C6H5-C'3HCHCH3 > > Me» SCH > CeHs-CHCH(CH3)2 CH3 05 On reduction of the acetates produced, 1-phenyl-2- methyl-l-propanol was obtained in amounts of 32% (of total alcohol formed) from th£ggfamine, 6% from erythro-amine. In the thggg case, 16% of this alcohol was found to be 0p- tically active; in the erythro only 3%. Cram and McCarty advanced a scheme (see sequence 7) that could account for the activity of the methyl-rearranged H CH CH 3 // I?! ,CE'IE! \\\ \ I} / + \ SOH CGHPC-CINCHQ. )2 ’/C*C\ _> ~> H__-C:._--\C- CH > I H’ ’ ~ 3 SO C6H5/ NH2 C6H5 \H ' active 11 7 1 ( ) + C6H5 "CHCH(CH3 )2 SOH> C6H5$HCH(CH3 )2 OS racemic product. Its important feature was an asymmetric methyl— bridged cationic intermediate (11) that either was captured 10 by solvent at C-1 to give active acetate, or opened up af- fording a symmetric 1-pheny1-2-methylpropyl cation that yielded racemic acetate. These authors acknowledged that another mechanism could be formulated in which the bridged ion ll'is a transitory intermediate. The difference in optical activity in the thggg and erythro runs could be ascribed to different rates of capture of sterically differ- ent disolvated ions arising from the respective bridged transi- tion states. In 1960 Silver (26) disclosed more evidence supporting a cyclic cationic intermediate. Deamination of 3-methyl—2~ butylamine in acetic acid gave a hydrocarbon fraction con— taining 5.6% cis- and 10.1% trans-1,2-dimethylcyclopropane. + 7H3 (EH3 ? CH3CH-——CHCH3 CH3CHCHCH3 —> -——> CH3CHCHCH3 '> \ / l + CH3 NH2 1 1,2-H~ 1 .3 CH3 CH3 + CH3CH———CHCH3 emacncrIch2 —> ---> CH3CHCH2CH2 cré NH2 Interestingly, isoamylamine also afforded the 1,2-dimethyl- cyclopropanes (0.5% and 1%, respectively), but no 1,1-di— twathylcyclopropane. Silver later preferred (27) to use classical carbonium ions to account for the open-chain hydro- carbons observed, and found it difficult to reconcile the dimethylcyclopropanes produced with results that indicated a lac*:of methyl-bridging in the propyl system (12,13). 11 More recently, Bayless, Mendicino and Friedman (28) communicated their finding methylcyclopropane, which was attributed to hydrogen-bridged intermediates (lg'and 13), as a hydrocarbon product of the aprotic deamination of 1- butylamine, 2—butylamine and isobutylamine (see eq. 8). H" " ‘CHZ + \ + / cmacnzcm2 CH2 —> \ / _H+ / CH3tH———CH2 \ CH2 + % lg CH3CH2CHCH3 + CH3CH——-CH2 (8) -H C\HZ‘~~H / (EH3 \ t/ \ CH3CHCH2 —-—> CH3CH——C’Hz + 3.3:, Protonated cyclopropanes have also been invoked in mechanisms of deaminations of alicyclic systems. Edwards and Lesage (29) suggested 14,and/or 15,as possible formu- lations for intermediates in the deaminative reaction of 2—aminocyclohexanone. O o O NH2 —9 —-) l ‘9 + 44 H ‘A 14 15 rw rw Similar reactions observed by these authors are shown belxnv. NH2 0’ -+ —> O * C2 “8”) 12 L L 0 e ‘. —-> —> l. o (R f. 29) NH2 0 o NHz __> —-> +482 (Ref. 30) Intermediacy of bridged-hydrogen or bridged—carbon ions in the rearrangement of strained systems (g;q., neopentyl or related systems) has not been the subject of the degree of controversy which characterized the less—substituted alipha- tic systems. Early work by Winstein and Marshall (31) and by Bartlett (32) on the solvolyses of neopentyl and tri-tfbutyl deriva- tives, respectively, gave indications that alkyl participa- tion (as in 16) was assisting the ionization process. How— ever, Bartlett and Swain (33) again cautioned in 1955 that extreme reactivity of such highly branched derivatives might be due to relief of steric strain, as well as to participa- tion. Numerous searches for ions such as lg'were unsuccessful. For example, Winstein and Morse (34) found "no support for 13 bridged structures" in the rearrangement of a-phenylneopentyl derivatives. Brown and coworkers (35) Showed that carbon- ,Qfia / +\ ?H ’—_-\ - (CH3)2C-———CHC6H5 ,-—-x —> (CH3)2C-C'IHCH3 (fl-13 m C6H5 CH3CCH(C6H5 )x (CH3 )3C-CHC6H5 > (CH3 )zc—CHCHa C6H5 bridged intermediates did not play important roles in either the 2,2,3-trimethyl-3-pentyl or 2,3,3-trimethyl-2-pentyl systems in the chlorination of the reSpective alcohols or solvolysis of the chlorides. After studying rearrangement reactions of 2,3,3-trimethyl—2-chlorobutane-1-14C, Roberts and Yancey (36) concluded that the nonclassical ion 11' was less stable than the classical ions and that the ion 11 represented an energy maximum. The investigation by Karabatsos (CH3 )2c’_—__-__-_:\_c (CH3 )2 17 NV and Graham of the neopentyl—l-13C system Similarly proved that protonated cyclopropanes were not major intermediates in the rearrangement of that system (37). Silver (38), on finding that olefins produced during solvolyses of neopentyl and typentyl derivatives had a common precursor, excluded the intervention of a methyl-bridged carbonium ion such as 18; Finally, Karabatsos, Orzech and Meyerson (39) showed that the neopentyl deaminative rearrangement took place with no detectable 1,3—hydride shifts and without the intermediacy of protonated cyc10propanes or hydrogen-bridged ions. Thus, deamination of neopentylamines labeled (with 13C or CD2) at C-1 led to tfpentyl alcohols with all of the label at C-3 of the Efpentyl alcohol. CH3 CH3 + (CH3)2C-—13CH2X or (CH3)2C—13CH2+ -—~> (CH3)2C-13CH2CH3 ——> OH (CH3)2<‘:13CHZCH3 From the forgoing discussion it is seen that inter- mediacy of bridged ions in reactions of 1-propyl derivatives is evidenced by isotope-position rearrangement of the carbon skeleton and by formation of cyclopropane. The latter is observed not only in deamination reactions, but also in the deoxidamjon of 1—propanol (10). Disubstitution of the 1- propyl system at C-2 gives, on the other hand, neopentyl- related derivatives that are characterized by not undergoing deamination reactions gig bridged-ion intermediates. More- over, cyclopropanes were not detected in the deamination of neopentylamine (10,26) or the deoxidemjon of neOpentyl alcohol (10,40). Aliphatic derivatives which are structurally intermediate 15 between l—propyl and neopentyl undergo rearrangements which are less easily particularized. Thus, although a small amount of methylcyclopropane formation accompanies deamina- tion of 1-butylamines, 2-butylamines and isobutylamines (28) and deoxflifimion of the respective alcohols (10,41), isotope- position rearrangement in these and Similar systems is less documented. Accordingly, an investigation was undertaken, the object of which was to assess the effects on deamination reactions of methyl substitution at C-1 and C-3. To this end, isotopically-labeled 1- and 2-butylamines were pre- pared and deaminated under standard conditions. Analysis of product alcohols served as the basis for conclusions derived from this study. RESULTS AND DISCUSSION I. General The primary objective of this study was to determine the relative importance of mechanistic pathways considered by previous investigators to have possible Significance in the reactions of carbonium ions. Specificially, rearrange- ments of cations derived from the deaminations of 1- and 2-butylamines were examined in order to determine which, if any, of these pathways predominate and how they relate to reactions of similar aliphatic syStemS. The perchloric acid salts of three 1-butylamines and two 2-butylamines, isotopically labeled with deuterium at various positions, were prepared for deamination. 1—Butyl- 1,1—g2-amine was prepared directly from butyronitrile by reduction with lithium aluminum deuteride. 1-Butyl—2,2- gz-amine and 1-butyl-3,3—§2-amine resulted from lithium aluminum hydride reduction of the respective dideuterated nitriles, which were prepared by appropriate malonic ester syntheses shown in Figures 1 and 2 and described in detail in the EXPERIMENTAL. Lithium aluminum deuteride reduction of 2-butanone oxime gave 2-butyl-2-dfamine. Synthesis of 2—butyl-3,3-§2- amine was accomplished by way of 2-butanol-3,3-§2, as is schematically Shown in Figure 3. 16 17 CH3CH2BI KOH CH2(COOEt)2 _ > caacnzcmcoost)2 > CH3CH2CH(COOH)2 OEt . D20 150° CH3CH2CH(COOH)2 EIGHSHEE’ CH3CH2CD(COOD)2 -c02> CH3CH2CD2COOD 1.Soc12 soc12 CH3CH2CD2COOD §—§§:——> cnacnzcozcoqu 1653—» CH3CH2CD2CN + H20 1. LiAlH4 CH3CH2CD2CN _ > CH3CH2CD2CH2NH2 2 0 H20 I OH Figure 1. Synthesis of 1-butyl-2,2-g2-amine. CHaBr CH2(COOEt)z-———:—> CH3CH(COOEt)2 OEt KOH > CH3CH(c00H)2 D20 140° 2 Exchange» CH3CD(COOD)2 CH3CH(COOH) > cnacnzcoon + co2 1. LiAlH4 1. PBr3 CH3CD2COOD _> CH3CD2CH20H §——E35—> cascnzcnzsr 2.H20, on ' 2 NaCN 1. LiAlH4 CH3CD2CH2Br > CH3CD2CH2CN —e» > CH3CD2CH2CH2NH2 aq. CH30H 2.1.120, OH- Figure 2. Synthesis of 1—butyl-3,3-§2-amine. 18 1. LiAlD4 1. PBr3 (CH3CO)20 _> cnacozon -———————> CH3CDzBr 2. H20, OH 2. H20 1. Mg,ether TSCl caacnzsr 2 CH CHO > CH3CD2CHOHCH3 Pyridine> CH3CD2CHCH3 ' 3 TS NaN3 1 o LiA1H4 CH3CD2CHCH3 aq. EtOH> CH3CD2CHCH3 2. H20! OH-—> cnacnzcncn3 0T5 N3 NHZ Figure 3. Synthesis of 2-butyl-3,3-§2-amine. 19 Deaminations of the perchlorates were carried out ac- cording to the procedure of Roberts and Halmann (5). The product alcohols arising from the l-butylamines were col— lected and shown to be comprised only of 1- and 2-butanol, in agreement with earlier observations (7,42,43). The 2- butylamines afforded 2-butanol as the only alcoholic product, as previously noted (28). Trimethylsilyl ether derivatives (44) of each alcohol were prepared, collected and analyzed by mass specrometry. In some cases alcohols, as well as the trimethylsilyl ether derivatives, were also analyzed by nuclear magnetic resonance (nmr) Spectroscopy. Duplicate deaminations gave identical results. II. Mass Spectral Data Trimethylsilyl ethers of unlabeled 1? and 2-butanol were prepared. Mass Spectral data of these compounds ap- pear in Tables I and II. Since the experimental approach of the problem depended on the assumption that the alcohols were irreversibly formed, it was imperative to demonstrate that the product alcohols were stable to the reaction medium. An authentic, deuterated alcohol, 2-butanol-2-d, was prepared and subjected to the experimental conditions. Mass spectral data of the tri— methylsilyl ether derivatives of the authentic 2-butanol- 2-d_and of the 2-butanol isolated from the blank reaction are given in Tables III and IV. The Spectra are identical. 2O Table I. Mass Spectrum of CH3CH2CH2CHZOSi(CH3)3 M/e Pk. Ht. Mono. 136 0.1 0.0 135 0.7 0.6 134 6.6 0.8 133 99.9 -0.5 132 284.0 -3.2 (p - Me)+ 131 2424.0 2423.0 99.6% at mass 131 130 2.0 1.1 129 6.9 6.8 (C6H15081)+ 128 1.2 1.2 11.85, 4.13, 0.24 2 2432.1 106 1.1 0.1 105 22.6 0.7 104 55.7 0.3 (P - Pr)+ 103 567.0 562.5 88.2% at mass 103 102 36.9 33.7 101 32.7 32.2 (C4H7OSi)+ 100 1.3 0.3 9.61, 3.86, 0.17 99 9.3 9.3 2 638.0 147 14.0 Conc. HMDS = 0.28 vol. % 21 Table II . Mass spectrum of CH3CH2 CH (CH3 )OSi(CH3)3 M/e Pk. H-t. Mono. 136 0.2 0.1 135 0.7 -- 134 3.9 1.4 133 33.6 -0.4 132 78.7 -0.15 (p - Me)+ 131 660.0 640.9 98.8% at mass 131 130 3.0 2.2 129 5.3 5.0 (0.9315031)+ 128 0.9 0.9 11.85, 4.137 0.24 2 649.0 121 0.2 0.05 120 3.9 0.3 119 67.2 0.4 118 177.9 -0.5 (p - Et)+ 117 1662.0 1656.1 98.9% at mass 117 116 5.9 3.3 115 16.6 13.3 114 0.6 0.1 (c5H13081)+ 113 3.3 2.0 10.72, 3.96, 0.21 2 1674.8 147 403.0 Conc. HMDS = 7.4 vol. % 22 Table 111. Mass Spectrum of authentic CH3CH2CD(CH3)OSi(CH3)3 M/e Pk. Ht. Mono. 136 0.1 0.1 135 2.0 0.2 (p - Me)+ 134 29.0 -0.4 Mono. Dist'n. 133 85.8 1.6 98.8% of 1.6 0.2% g2 132 708.0 707.2 2 = 712.0 707.2 99.3%5;1 131 6.6 6.2 3.2 0.5%go 130 2.1 1.7 712.0 129 3.0 2.9 128 1.0 1.0 2 720.6 + 121 3.9 0.1 ,jP - Et) _ 120 71.0 -0.2 Mono. Dist'n. 119 193.2 0.4 98.9% of 0.4 -- 118 1794.0 1791.4 2 = 1808.3 1791.4 99.1%gL1 117 20.9 19.4 16.5 0.9%910 116 12.9 12.6 1808.3 115 2.8 2.7 114 1.0 0.9 113 1.0 1.0 2 1828.4 147 3.0 Conc. HMDS 3 0.06 vol. % 23 Table IV. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 2-butanol recovered from blank deamination with 2-butanol—21i. M/e Pk. Ht” Mono. 136 0.3 0.3 135 2.0 0.2 1p - Me)+ 134 29.9 -0.5 Mano. Dist'n. 133 87.8 '""BTZ" 98.8% of “"072’ 'TEEEEE' 132 736.0 735.2 2 = 738.4 735.2 99.6%g1 131 6.3 6.0 2.8 0.4969”o 130 2.0 1.6 ‘73874 129 3.4 3.3 128 0.9 0.9 2 747.4 122 0.1 0.6 121 4.3 0.4 120 74.0 —0.3 (p - 2t)+ 119 200.7 —1.1 Mono. Dist'n. 118 1878.0 1875.5 98.9% of 1875.5 99.2%g_1 117 19.6 18.1 2 = 1889.9 14.4 0.8%5;2 116 13.4 13.1 'I889T9 115 2.6 2.5 114 0.9 0.8 113 0.9 0.9 2 1910.9 147 1.0 Conc. HMDS = 0.02 vol. % 24 In Tables V and VI, respectively, are Shown the mass Spectra of the derivatives of the 1- and 2-butanols isolated from the deamination product mixture of 1-butyl-1,1fg2- ammonium perchlorate. The mass spectra of authentic 1-butyl- 1,1-Q3 trimethylsilyl ether and 2-butyl-1,1-d3 trimethylsilyl ether are presented in Tables VII and VIII, respectively. Trimethylsilyl ethers derived from 1- and 2—butanols resulting from the deamination of 1-butyl-2,2fi§2-ammonium perchlorate were prepared and their mass Spectra are Shown in Tables IX and X. Tabulation of the mass spectrum of the ether of authentic 1—butanol-2,2-§2 is given in Table XI. Mass Spectra of the derivatives from 1- and 2-butanol collected from the deamination of 1-butyl-3,3-§2-ammonium perchlorate are detailed in Tables XII and XIII. The mass Spectrum of authentic 1-butyl-3,3-g2 trimethylsilyl ether is shown in Table XIV. The mass spectrum of the 2-butanol obtained by de- aminating 2-butyl-2-gfammonium perchlorate is displayed in Table XV. Mass spectral data of authentic 2—butyl—24d trimethylsilyl ether have already been given (Table III). Tables XVI and XVII relate to the 2-butyl-3,3-dz sys- tem. The former Shows the mass Spectrum of the ether of the 2-butanol obtained from the deamination while the latter details the mass Spectrum of authentic 2-butyl—3,3-d3 tri— methylsilyl ether. 25 Table V. Mass spectrum of CH3CH2CH2CH20Si(CH3)3 from 1-butanol obtained from the deamination of 1-buty1-1,1-d2-ammonium perchlorate. M/e Pk. Ht Mono. 137 0.2 0.2 136 5.1 0.1 135 84.9 —1.7 (P - Me)+ 134 246.6 -3.2 Mono. Dist'n. 133 2100.0 2095.6 99.7% of 2095.6 93-3%.§2 132 35.3 34.6 2 = 2131.8 34.6 1.6% d1 131 5.7 5.1 1.6 0.1%30 130 2.7 2.7 2131.8 129 0.2 0.7 2 2138.2 107 17.3 0.0 (P — Pr)+ 106 43.4 -0.9 Mono. Dist'n. 105 450 .0 445 .9 88 .1% of 445 .9 97 .9% <_i_2 104 37.4 35.6 2 = 455.4 9.5 2.1% d1 103 16.2 15.0 455.4 102 10.1 9.5 101 5.1 4.6 100 3.0 2.6 99 3.7 3.7 2 516.9 147 5.9 Conc. HMDS = 0.12 vol. % 26 Table VI. Mass spectrum of CH3CH2CH(CH3)OSi(CH3)3 from 1—butanol obtained from the deamination of 1-butyl-1,1-d2-ammonium perchlorate. M/e Pk. Ht. Mono. 136 1.3 0.0 135 21.4 -0. (P - Mel+ 134 62.6 -0.1 Mono. Dist'n. 133 529.0 522.8 98.8% of 522.8 82-7%.§2 132 23.0 10.3 2 = 632.2 10.3 1.6% d1 131 107.1 105.5 99.1 15.7% 9.0 130 1.1 1.1 632.2 129 0.2 0.2 2 639.9 122 2.8 0.2 121 49.2 0.2 (P - 2t)+ 120 132.6 -1.7 Mono. Dist'n. 119 1254.0 1236.0 98.9% of 1236.0 75.6% d2 118 66.9 25.8 2 = 1633.8 25.8 1.6% d1 117 382.0 381.1 372.0 22.8% do 116 3.5 2.7 1633.8 115 6.2 5.8 114 0.5 0.5 113 0.2 0.1 2 1652.0 147 33.9 Conc. HMDS = 0.68 vol. % 27 Table VII. Mass Spectrum of authentic CH3CH2CH2CDZOSi(CH5)§ M/e Pk. Ht. Mono. 136 6.3 0.3 135 100.8 -1.8 (P - Me)+ 134 292.0 -3.6 Mono. Dist'n. 133 2487.0 2481.9 99.6% of 2481.9 98.1% 512 132 38.8 36.6 2 = 2529.7 36.6 1.5% g, 131 18.0 17.4 11.2 0.4% go(?) 130 3.2 3.1 2529.7 ' 129 0.3 0.2 128 0.7 0.7 2 2539 9 108 0.8 -0.1 + 107 20.5 0.1 (P- Pr) 106 51.7 —0.7 Mono. Dist'n. 105 532.0 526.9 88.2% of 526.9 96.7% d2 104 46.8 44.3 2 = 544.7 17.8 3.3%.511 103 22.6 21.1 544.7 102 13.0 12.3 101 6.2 5.8 100 3.1 2.7 99 4.5 4.5 2 617.6 147 4.8 Conc. HMDS = 0.10 vol. % 28 Table VIII. Mass spectrum of authentic CH3CHZCH(CHDZ)081(CH§)5 M/e Pk. Ht. Mono. 136 1.1 0.0 135 19.9 -0.4 (P rue)+ 134 58.6 0.7 Mono. Dist'n. 133 489.0 481.8 98.8% of 481.8 77.4%g2 132 26.4 10.0 2 = 622.2 10.0 1.6%g_1 131 137.7 136.5 130.4 21.0%g0 130 1.2 1.1 622.2 129 0.4 0.4 2 629.8 122 3.8 0.3 121 63.6 -0.5 (P - at)? 120 176.4 3.7 Mono. Dist'n. 119 1611.0 1607.1 98.9% of 1607.1 98.1% g2 118 33.1 32.1 2 = 1638.7 31.6 1.9%31 117 9.8 9.1 1638.7 116 2.0 1.1 115 7.2 7.0 114 0.4 0.4 113 0.2 0.1 2 1656.9 147 23.0 Conc. HMDS = 0.49 vol. % 29 Table IX. Mass spectrum of CH3CH2CH2CHZOSi(CH3)3 from 1-butanol obtained from the deamination of 1-butyl-2,2-g2-ammonium perchlorate. M/e Pk. Ht. Mono. 137 0.2 0.1 136 5.9 0.6 135 89.0 -2.4 (P - Me)+ 134 260.7 -4.7 Mono. Dist'n. 133 2220.0 2209.4 99.6% of 2209.4 96.3% 512 132 86.6 85.9 2 = 2293.8 84.4 3.7% d1 131 5.5 4.4 2293.8 130 2.3 2.2 129 0.3 0.2 128 0.9 0.9 2 2303.0 + 107 0.2 -0.1 (P - Pr) 106 3.6 1.3 Mono. Dist'n. 105 22.8 3.3 88.2% of 3.3 0.9% d2 104 71.7 30.1 2 =430.4 30.1 7.0%g_1 103 430.0 428.5 397.0 92.2% do 102 12.9 11.8 430.4 101 10.0 9.6 100 3.1 2.8 99 1.9 1.9 2 488.0 147 12.2 Conc. HMDS = 0.24 vol. % 30 Table X. Mass Spectrum of CH3CH2CH(CH3)OS1(CH3)3 from 2-butanol obtained from the deamination of 1-butyl-2,2—dz-ammonium perchlorate. M/e Pk. Ht. Mono. 136 1.7 0.3 135 24.0 —0.1 (P - Me)+ 134 73.0 -0.5 Mono. Dist'n. 133 590.0 573.6 98.8% of 573.6 81.0% d2 132 135.6 134.8 2 = 708.4 134.8 19.0% d1 131 5.5 4.8 708.4 130 2.6 2.5 129 0.4 0.3 128 1.0 1.0 2 717.0 122 3.1 0.4 121 51.4 0.3 (P - 2t)+ 120 140.4 -1.1 Mono. Dist'n. 119 1311.0 1286.1 98.9% of 1286.1 72.0% d2 118 117.0 69.6 2 = 1786.0 69.6 3.9% Q1 117 440 .0 439 .0 430 .3 24 . 1% do 115 8.7 8.5 1786.0 115 1.7 1.5 114 1.0 1.0 113 0.2 0.2 2 1805.9 147 8.9 Conc. HMDS = 0.18 vol. % 31 Table XI. Mass spectrum of authentic CH3CH2CD2CH2081(CH§)§ M/e Pk. Ht. Mono. 136 7.1 0.9 135 105.0 2.0 (P - Me)+ 134 298.0 0.05 Mono. Dist'n. 133 2499.0 2490.3 99.6% of 2490.3 97.3% g2 132 70.2 68.6 2 = 2560.2 68.6 2.7% d1 131 9.0 8.1 1.3 trace dB 130 2.5 2.4 2560.2 129 1.0 1.0 2 2570.5 + (P - Pr) 106 2.9 -0.6 Mono. Dist'n. 105 30.2 8.4 88.2% of 8.4 1.7% d2 104 81.0 34.7 2 = 483.2 34.7 7.2% d1 103 478.0 476.3 440.1 91°1%.Qo 102 14.1 12.9 483.2 101 11.2 10.8 100 3.0 2.7 99 2.0 2.0 2 547.8 147 13.4 Conc. HMDS = 0.27 vol. % 32 Table XII. Mass Spectrum of CH3CH2CH2CHZOSi(CH3)3 from 1-butanol obtained from the deamination of 1-butyl—3,3-§2-ammonium perchlorate. M/e Pk. Ht. Mono. 137 0.3 0.3 136 5.5 0.2 135 88.0 -3.7 (P - Me)+ 134 259.8 -6.8 Mono. Dist'n. 133 2226.0 2214.7 99.6% of 2214.7 95-7%.§2 132 100.8 100.0 2 = 2313.4 98.7 4.3%g_1 131 6.4 5.7 2313.4 130 1.2 1.1 129 0.3 0.2 128 1.0 1.0 2 2322.7 (2- Pr)+ 106 1.1 0.1 Mono. Dist'n. 105 20.2 0.9 88.2% of 0.9 0.2% _cl_2 104 50.5 1.9 2 = 493.6 1.9 0.4%g_1 103 496.0 491.7 490.8 99.4%go 102 38.3 35.8 493.6 101 23.8 23.3 100 3.8 3.4 99 2.7 2.7 2 559.7 147 13.0 Conc. HMDS ‘ 0.25 vol. % 33 Table XIII. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)§ from 2-butanol obtained from the deamination of 1—butyl-3,3-§2-ammonium perchlorate. M/e Pk. Ht. Mono. 136 2.1 0.3 135 30.9 -0.6 (P - Me)+ 134 89.7 -1.9 Mono. Dist'n. 133 766.0 761.9 98.8% Of 761.9 96.3% 512 132 32.0 31.4 2 = 791.5 29.6 3.7% $11 131 4.2 3.0 791.5 130 3.2 3.1 129 0.8 0.7 128 1.0 1.0 2 801.1 121 0.7 -0.1 120 18.0 0.5 (P - 2t)+ 119 99.3 -0.3 Mono. Dist'n. 118 526.0 361.2 98.9% of 361.2 19.2%‘Q1 117 1536.0 1535.1 2 = 1883.8 1522.6 80.8%‘Q0 116 6.8 6.5 1883.8 115 1.3 1.1 114 0.9 0.9 113 0.1 0.0 2 1904.8 147 16.6 Conc. HMDS = 0.32 vol. % 34 Table XIV. Mass spectrum of authentic CH3CD2CH2CH20Si(CH5)3 M/e Pk . H-t. Mono . 137 0.2 0.2 136 5.0 0.3 135 80.0 -1.6 (P - Me)+ 134 233.1 -3.4 Mono. Dist'n. 133 1980.0 1970.8 99.6% of 1970.8 96.5% $2 132 73.7 72.9 2 = 2043.1 72.3 3.5% d1 131 6.4 5.6 2043.1 130 1.0 1.0 129 0.2 0.1 128 0.9 0.9 2 2051.3 (P - Pr)+ 106 0.8 -0.1 Mono. Dist'n. 105 18.3 1.4 88.2% of 1.4 0.3% $2 104 44.6 2.1 2 = 431.6 2.1 0.5% Q; 103 433.0 429.2 428.1 99.2%go 102 33.7 31.6 431.6 101 20.6 20.2 100 3.0 2.7 99 2.1 2.1 _______2 2 489.3 147 14.4 Conc. HMDS = 0.29 vol. % 35 Table XV. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)§ from 2-butanol obtained from the deamination of 2-butyl-24Q-ammonium perchlorate. M/e Pk. Ht. Mono. 135 2.9 0.7 ((P - Me)+ 134 26.9 2.2 Mono. Dist'n. ._______, .______ .________ 133 80.1 13.7 98.8% of 13.7 2.4% 912(2)) 132 558.0 557.0 2 = 574.3 557.0 97.0% Q1" 131 7.0 6.3 3.6 0.6% 510 130 2.3 2.3 574.3 129 2.0 2.0 2 581.3 122 0.5 0.2 121 4.9 0.8 (P - Etf' 120 56.1 2.6 Mono. Dist'n. 119 171.0 28.3 98.9% of 28.3 1.9% d3(?) 118 1284.0 1264.3 2 = 1469.5 1264.3 86.1% Qt ' 117 181.5 180.2 176.9 12.0% 90 116 10.4 10.2 1469.5 115 1.9 1.7 114 0.4 .3 113 0.8 0.8 2 1485.8 147 7.0 Conc. HMDS = 0.15 vol. % 36 Table XVI. Mass Spectrum of CH3CH2CH(CH3)OSi(CH3)§ from 2-butanol obtained from the deamination of 1—butyl-3,3-g2-ammonium perchlorate. M/e Pk. Ht, Mono. 136 1.8 0.1 135 28.0 -0.3 (P — Me)+ 134 81.9 -0.1 Mono. Dist'n. 133 690.0 678.8 98.8% of 687.8 97.5% d2 132 13.2 11.7 2 = 705.6 11.7 1.6% d1 131 12.3 11.5 6.1 0.9% dB 130 2.0 1.9 705.6 129 0.6 0.5 128 0.8 0.8 2 714.2 + 121 0.6 0.2 (P - Et) 120 9.4 -0.2 Mono. Dist'n. 119 77.8 0.7 98.9% of 0.7 0.1% fig 118 327.0 159.1 2 = 1667.8 159.1 9.5% d1 117 1518.0 1517.1 1508.0 90.4% 90 116 6.3 6.1 1667.8 115 1.7 1.5 114 1.2 1.1 113 0.7 0.7 2 1686.3 147 14.0 Conc. HMDS = 0.29 vol. % 37 Table XVII. Mass spectrum of authentic CH3CD2CH(CH3)OSi(CH3)3 M/e Pk. Ht” Mono. 136 2.1 0.1 135 33.0 0.0 (P -Me)+ 134 94.8 -0.5 Mono. Dist'n. 133 801.0 798.4 98.8% of 798.4 97.4% d3 132 16.0 14.2 2 = 819.6 14.2 1.7%_c_1_1 131 14.0 12.7 7.0 0.9%g_O 130 2.5 2.3 819.6 129 1.0 0.9 128 1.1 1.1 2 829 6 121 1.0 0 9 (Pg- 2t)+ 120 4.7 0.4 Mono. Dist'n. 119 78.3 1.2 98.9% of 1.2 0.1% _d_2 118 209.7 1.3 2 = 1932.5 1.3 0.1%.g1 117 1941.0 1939.8 1930.0 99°8%.§o 116 7.9 7.5 1932.5 115 2.2 1.8 114 1.8 1.7 113 0.8 0.7 ___—_——J 2 1954 0 147 22.0 Conc. HMDS = 0.45 vol. % 38 III. 1-Butyl Systems Although acknowledgment is made of formation of other (particularly hydrocarbon) products (7,28,43), the relevance of which to the present study is discussed later, this investigation was chiefly concerned with the alcohols pro- duced. That these were formed irreversibly was illustrated by recovering 2-butanol-2—g_unchanged after subjecting it to the reaction conditions. In Tables XVIII and XIX are compiled mass Spectral isotopic distributions of the Silyl ethers of all authentic and product 1- and 2—butanols, in- cluding the tested alcohol before and after the blank de- amination. Shown in Figure 4 are carbonium ion rearrangement mechanisms which have been considered by previous investi- gators in related systems. These, as applied to the 1-butyl model, include (a) 1,2-hydride shifts, (b) 1,3-hydride Shifts, (c) 1,4-hydride Shifts, (d) 1,2—ethyl Shifts, (e) 1,3-methyl Shifts and (f) involvement of bridged ions or protonated cyclopropanes. Of course, combinations of pathways would lead to other degrees of scrambling of the carbon or hydro— gen atoms. Possible competition of reaction modes between carbonium ions in the system and precursors of carbonium ions, such as alkyl diazonium ions (42,45,46), also must be recognized. ~Although Figure 4 depicts how, by reasonable mechanisms, a 1-butyl cation (or its precursor in the deamination of 1- butylamine) might be converted into derivatives of isotope- 39 Table XVIII. Results of mass Spectral analyses of the trimethyl- silyl ethers of authentic, deuterated alcOholS. CH3CH2CH20H2081(CH3)3 CHBCHZCH(CH3)081(083)5 + + + ‘ + (P-CH3) (P-C3H7) (P-CH3) (P-C2H5) Authentic 98.1%‘d2 96-7%.Qz CH3CH2CH2CDZOSi(CH3)3 1.5%g1 3.3%9_1 0.4%_c_1_0 Authentic 97.3% $3 1.7% Q2 CH3CH2CD2CH20Si(CH3)3 2.77531 7.2% d1 trace g” 91.1% do Authentic 96.5% $2 0.3% Q2 CH3CD2CH2CHZOSi(CH3)3 3.5%91 0.5%g1 99.2% 530 Authentic 0.2% Q2 CH3CH2CD(CH3)OS1(CH3)3 99.3%31 99.1%g_1 05% 5.1.0 03% 5.3.0 Authentic 77.4% g2 98.1% d2 CH3CH2CH(CHD2)OSi(CH3)3 1.6%31 .1.9%.gl 21.0%;gb Authentic 97.4% Q2 0.1% g2 CH3CD2CH(CH3)OSi(CH3)3 1.7%31 0.1%g_1 0.9% go 99.8% 910 40 Table XIX. Results of mass Spectral analyses of trimethyl— silyl ethers of deamination product alcohols. CH3CH2CH2CH20S1(CH3)3 CH3CHZCH(CH3)OSi(CH5)3 (P-CH3)+ (P-c3H7)+ (P-CH3)+ (P-C2H5)+ From CH3CH2CH2CD2NH2 98.3% g2 97.9% g2 82.7% 512 75.6% £12 1.6% gl 2.1% gl 1.6% d1 1.6% _c_1_1 0.1% go 15.7% dO 22.8% go From CH3CH2CD2CH2NH2 96.3% (33 0.8% 512 81.0% 22 72.0% 512 3.7% g, 7.0% 91 19.0% 511 3.9% 311 92.2% go 24.1% go From CH3CD2CH2CH2NH2 95.7% 92 0.2% _c_1_2 96.3% 532 4.3% 91 0.4% 911 3.7% gl 19.2% 311 99.4% 510 80.8% _d_0 From CH3CH2CD(CH3)NH2 2.4% 92(?)1-9% 151.2(2) 97.0% g, 86.1% 61‘ 0.6% 510 12.0% 9,, From CH3CD2CH(CH3)NH2 97.5% 512 04% 22 1.6% g, 9.5% g, 0.9% go 90.4% go From blank deamination (No 1-butanol trace g2 detected) with CH3CH2CD(CH3)OH 99.6% 91 994% 91.1 0.4% go 0.8% 510 41 etc. etc. H H + + CH3CH2CH2CH2 CHZCHZCHZCH3 H H + + CH3CH2CHCH3 9:3) CH3CHCH2CH3 /H\ I, + \\ + CH3-CH CH2 <:> CH3CHCH2CH3 A CH2 A 1,2-H~ /H\ 1’ + \\ + CH2 CH2 <:>CH2CH2CH2CH3 F" '1 éHz— CH2 CH3CH2CH2CH2X or CH3 CH CH CH CH ~47 CH __ 3 2 2 2 J 1,2—Et~ + . 2 . CH2CH2 F \\\\\; E p _+ CH3 1'3'Me“' CHZCHZCHZ (r 93 93 93 [egg Q§‘~ CH // + \ and/or \ + [H __> / \ \ <__._.. CHz—q-CHz CH2-—-CH2 CH2--7pH2 \ +/ \ / H H 913 + CH 1‘ 9H3 CH3 CH2__'CH3 CH 1.2—H~' +C ‘————* ,/ \\ ' CH + CH3 3 CH2 CH3 Figure 4. Possible rearrangement pathways of the 1-buty1 cation. 42 position rearranged l—butyl, 2-butyl, isobutyl, and even £— butyl (by way of isobutyl) functions, some of the pathways can be, and have been, excluded. Thus, the observation that isobutyl and tfbutyl solvolysis products are not formed in the deamination of l—butylamine (7,28,47) reduces the Signi- ficance of mechanisms involving isobutyl and Eybutyl cations as intermediates. Study of the isotopic distribution of the 1- and 2*butanols formed from the deamination of suit- ably labeled 1—butylamines allows exclusion of some other pathways of Figure 4. A. 1-Butanol Formation The 1-butanols were found to have isotopic distributions nearly identical to those of the amines from which they were respectively produced. In regard to the 1-butyl-1,1-g2 system, this observation precludes the participation of re- arrangements C, D and E (in Figure 4 and below) as Signifi- cant in fireproduction of 1-butanol. Streitweiser and Schaeffer (42) have shown 1,2-ethyl shifts to be absent in this system. OH + . 1 i-H... Th CHZCH2C32CHD2~>CH2CH2CH2CHD2 F q CH3CH2CH2CD2X + OH or 1 3.31321“? CH2CD20H2€H3 -> CH2CD2CH2CH3 + I fH3CH2CH2CD2 ‘- OH 1 5541.3me CH2€H2CD2€H3 —> CH2CH2CD2CH3 43 The fact that the (P-CH3) ion g2 contribution of the ether of the l-butanol equals that of the ether of authentic l—butanol-1,1-d2 indicates that a mechanism involving diazo- alkane or carbene formation can account for no more than 1% of the product alcohol (6,11,28,42,48). Symmetrical bridged ions, such as 22x are excluded as I CH3CH2CH2CDZOH CH2 ,’ \ > + I + \\ CH2==:CDz CH3CH2CD2CH20H (not detected) CH3CH2CH2CD2XZ )x7 22. or + CH3CH2CH2CD2 J x CH3 CH3 — 19 i 4 CH 4H m \\ ‘ \ “~ / \\+ [H < > / \ j/D —> CH3CHDCH2CHDOH / cnz— c02 CH2— ‘CH0 (not a; 22 detected) intermediates (in the formation of 1—butanol) as 1-butanol- 2,2-g2 was not detected. Equilibration of ions such as 21 and 22 with subsequent formation of 1-butanol is likewise disallowed. Moreover, absence of deuterium scrambling in the 1-butanol resulting from the 1-butyl-1,1—g2 and 1-butyl- 2.2-92 runs disfavors any mechanism involving interconvert- ing primary and secondary cations. This observation sup— ports previous conclusions (5,13,49). CH3CH2CD2CH2X + + or + -——> CH3CH2CDCH2D 2:3.CH3CH2CHDCHD > CH3CH2CD2CH2 CH3CH2CHDCHDOH (not detected) 44 It can therefore be concluded that in the deamination of 1—butylamine, 1-butanol arises from displacement on a 1-butyl cation precursor by solvent and/or from capture by solvent of a 1-butyl cation. Strietweiser and Schaeffer (42) showed that at least 69 i 7% of the 1-butyl-1jd7acetate . obtained in the deamination of 1—butyl-1jdfamine in acetic acid resulted from displacement that inverts the configura- tion at C-1. B. 2-Butanol Formation From Table XIX, the isotopic compositions of the 2— butanols formed from 1-butyl-1,1-d2-amine (98.4% $2 and 1.6% gl) may be calculated. These are corrected by removing the contribution of the singly deuterated amine and shown below. The Q” (P-C2H5) ion is attributed solely to 22 with no contribution by CH3CHOHCHDCH2D or CH3CHOHCD2CH3, as mechanisms of interconverting primary and secondary cations have been excluded. OH OH CH3CH2CH2CD2NH2 > CH3CH26HCHD2 + CH3CHCH2CHD2 21 22. 100% 77% 23% It should be noted that the data of Table XIX indicate that the (P-CH3) ion should be 22.8% g2 if C-methyl frag— mentation of the 2-butyl trimethylsilyl ether occurred exclusively and 98.4% d2 if only Si-methyl cleavage occurred. Since the (P-CH3) ion was observed to be 82.7%, the contri- bution of (P-CH3) ion by loss from the trimethylsilyl group 45 is (82.7-22.8)/(98.4-22.8) x 100% = 79%, with 21% by loss from.the 2*butyl group. The mass Spectral data of the ether of authentic 2—butanol-1,1—d2 are in accord with the calculated values. 2-Butanol arising from 1-butyl-2,2-g2—amine (96.3% d2 and 3.7% gl) is calculated to be the following. (IDH OH CH3CH2CD2CH2NH2 ——> CH3CH2CDCH2D -F CH3CHC2H3D2 25 100% 75% 25% Finally, 1—butyl-3,3-§2—amine (95.7% Q2 and 4.3% Q1) afforded 2-butanol which, from the (P-C2H5) ion of its ether. is indicated to be as shown. OH OH OH CH3CD2CH2CH2NH2 ——> CH3CD26HCH3 + CH360CHDCH3 + CH3CHD6HCH2D 2.9 :42. 29. \\l~_’_,,__u\/S—-L~___,/ 100% 81% 19% It is possible to distinguish between mechanisms of formation of 2-butanol. From the distributions given above it is evident that the predominant path involves a 1,2- hydride (or deuteride) Shift from C-2 to C-1 (rearrangement path A in Figure 4). The lesser amount(S) of 2-butanol may possibly be derived from (1) subsequent 1,2-hydride shifts of the newly formed 2-butyl cation or (2) nominal 1,3-hydride shifts (paths B or F) of the 1—butyl cation or its precursor. 46 The alcohol 2g’(from 1—butyl-1,1—d2-amine) is consistent with both paths B and F cessive 1,2-shift mechanism. r- _. ~ + CHscHZCHZCDZX) 1'2 H > CH3CH2CHCHD2 or + 1,3- a. + CH3CH2CH2CD2 > CH3CHCH2CHD2 of Figure 4, as well as a suc- _.e + l-L--2-—§I-—> CH3CHCH2CHD2 t OH > CH3 CHCH2 CHD2 res. Although different 2—butanols (22 and 22) are formed 1 in the 1-butflr2,2—g2 system by the two proposed mechanisms, mass spectral data cannot distinguish them since the (P-C2H5) ion of each would be a d9 species, and the (P-CH3) ion a £2 Species. 1,2- ~ + — ~ + P - D > CH3CH2CDCH2D 142—§—> CH3CHCHDCH2D -> CH3CH2CD2CH2X ?H or CH3CHCHDCH2D + 22. LCH3CH2CD2CH2 J OH ' 1,3-H~ -+ l > CH3CHCD2CH3 > CH3CHCD2CH3 29 Also, the (P-C2H5) ion does not allow differentiation of the alternative (21 and 22) of the 1-butyl-3,3-d2 model. The (P-CH3) ion in formation is more combination with nmr data. in the 2-butyl cation compete success definitive, however, in If subsequent 1,2-hydride shifts fully with solvent capture, alcohol 22 might also be expected as a significant product. That it is not, that the (P-CH3) ion g1 contribution is supported by the observation of the 2-butanol (3.7%) 47 oH CH3CD2CHOHCH3 CH3CDCHDCH3 26 27 FEH CD CH CH X-Z/gdgtfit; CH CD CHCH I'Z’D“3 CH CDCHDHC 3 2 2 2 3 2 3"““- 3 3 or + 1 3—D~ '+ 1 2- ~ + CH3CD2CH2CH2 \—'——> CH3CDCH2CH2D ""-'—I-I—’> CH3CHDCHCH2D A 1 1 ?H ?H CH3CDCH2CH2D CH3CHDCHCH2D 2:2, 21, derived from the deamination is smaller than the Q; present in the starting material (4.3%). Having established the absence of §A'(i;g,, no deuterium at C-l), nmr permits differentiation of ggfl which contains deuterium at C-4, and 21, which does not. Under the spec- troscopic conditions used, the ratio of integrated signals of C—1 methyl (appearing at 18.87) to C-4 methyl (at T9.10) of the trimethylsilyl either of unlabeled Z-butanol was essentially the same as that for the ether of the derived 2-butanol (see Figure 5 and Table XX). As the predominant alcohol (gfi), arising from a single 1,2-hydride shift, also should contain no deuterium in either methyl group, equal ratios are indicative of all of the product 2-butanols hav- ing deuteria only at C-2 and C-3. Since the only detectable 2-butanols produced from 1- butyl-3,3-g2—amine have been shown to be the alcohols fig, and 21“ the 81:19 ratio of their concentrations is supported by careful integration of the nmr spectrum of the mixture 48 A n l. .12. ~ "A CH3CHCD3CH3 ~ OSl(CH3)3 + §_' ELL CH3CDCHDCH3 OSi(CH3)3 B + B' A. ....) | I H0 T 8.87 T 9.10 Figure 5. Methyl region of the nmr spectrum of the trimethyl- silyl ether of the 2-butanol obtained from 1-butyl- 3-3-d2-amine. 49 Table XX. Integrated nmr signals of methyl protons of the trimethylsilyl ethers of 2-butanol obtained from the deamination of 1-butyl-3,3—§2-ammonium per- chlorate and of unlabeled 2-butanol. Signal Compound Protons T Intensity Ratio sec—C4H7D208i(CH3)3 1-Methyl 8.86 7.1 1.08 4-Methyl 9.10 6.6 1.00 CH3CH2CH(CH2)Si(CH3>3 1-Methyl 8.86 7.8 1.07 4-Methyl 9.10 7.3 1.00 Table XXI. Integrated nmr signals of various protons of the 2-butanol obtained from the deamination of 1- butyl-3,3—§3-ammonium perchlorate. Signal . Proton(s) T Intensity Ratio Hydroxyl 5.23 16.1 1.00 Methine 6.35 13.1 0.81 l—Methyl triplet 8.87 .7 0.19 l-Methyl doublet 8.86 7.2 0.81 l-Methyl total -- 8.9 1.00 50 (see Figure 6). Thus, the hydroxyl protonzmethine proton ratio was found to be 1.0:0.81. Also, the l-methyl groups of 26 and 21 gave a doublet (J = 6.0 cps) and triplet (J = 0.9 cps), respectively, whose integrated signal ratio was 0.81:0.19 (see Table XXI). The 4—methyl groups did not give absorptions sufficiently resolved to permit separate integration. Therefore, all 2—butanol produced in deamin- ative reactions of 1-butylamine is accounted for by a series of 1,2—hydride shifts. No evidence supporting bridged-ion intermediates or 1,3—hydride shifts was obtained. IV. 2—Butyl Systems Figure 7 illustrates some rearrangement mechanisms which a 2-butyl cation might be expected to undergo. Because 2-butanol is the only alcoholic product of the deamination of 2-butylamine (28), mechanisms involving formation of 1-butyl (paths A2 and B, for example), isobutyl and E— butyl (path C) cations need not be considered as significant pathways. From what has been observed regarding the behavior of 2—butyl cations in deamination reactions of 1—butylamine, partial scrambling of C-2 and C—3 hydrogens is expected through a series of reversible 1,2-shifts. Determination of whether or not more complex rearrangements occur, by means of bridged—ion intermediates, requires closer examina— tion with the aid of isotOpic labeling. 51 .mcflEMImmrm.mIHmu5QIH Eonm Umcflmuno HocmudQIN mnu mo Eduuummm HEZ mw.m P 3.? mm.m p .m+m : cm T mo . mmoooomommo .4m .dm.wm + m 8 mmomomoommo ¢ Q m .m+m .m mnsmflm 7) 52 etc. etc. H H + + CH3CHCH2CH3 CH3CH2CH2CH2 A1 1”2—H~ H A2 / \ I \ + ’ + \ '— .., CH2 CHCH3 -""'> CH2CH2CH2CH3 X ' B CH CH3CH2CHCH3 2 1.3-H~ or + CHaCH2CHCHs ‘\\\“‘E~\~\, '+ $H3 1 2-H~. $H3 — ... _ ~ —.L_.___— 1,2 Me CH2CHCH3 > CH3ECH3 D I \3 ‘2 x‘ ’z I 2 I, + \ and/or \ t/H and/or lit, CH2: CH2—_—(':H CH2—_(':H CH3 CH3 CH3 Figure 7. Possible rearrangement pathways of the 2-butyl cation. 53 2-Butyl-2fid System. If 1,2-hydride (or deuteride) shifts were the only detectable rearrangements taking place, the 2-butanol formed from 2-butyl—2-g7amine would be a mix- ture of the two alcohols 32,and §§x neither of which have deuterium in a methyl group (see sequence 9). Whether or not there was deuterium at C-1 or C-4 was established by using nmr and mass spectral data. The ratios of integrated signals of 1-methyl protons to 4-methyl protons (Figure 8) _. I OH X l (E /——> CH3CH2 CDCH3 m + or CH3 CH2 CDCH3 + 1 , 2 -H~ ( 9 ) CH3CH2CDCH3 1 2_H~ + *- — ' CH3CHCHDCH3 1 2_H~ J \fl-w ' am \ . CH3CHCHDCH3 ~6— CH3CHDCHCH3 33 was found to be the same for the derived 2-butanol product(s) as for standard, unlabeled 2-butanol (see Table XXII). Thus, any deuterium present in the methyl groups must be equally distributed between C—1 and C-4 of various Z-butanol mole— cules. Since the (PbCHa) ion of the product 2-butanol lost essentilly no deuterium [0.6%.g0; (P-CH3) ion of authentic Zrbutanol-2—g was 0°5%.§0K50)' there could not have been a significant amount of deuterium migration into the 1-methyl group (of the final products). This implies no deuterium at C-4 since presumably the reversible sequence 10 would 54 §_._Q. EL. CH3CH2CDCH3 OH + liLSlL ELL CH3CHDCHCH3 0H B+B' C+C' AI H0 1' 8.87 Figure 8. Methylene and methyl regions of the nmr spectrum of the 2—butanol obtained from Z-butyl-Zegfamine. 55 Table XXII. Integrated nmr signals of various protons of the 2-butanol obtained from the deamination of 2-butyl-2-d-ammonium perchlorate and of un- labeled l-butanol. Compound Protons T Intensity Ratios sec-C4H8DOH 1-Methyl triplet 8.87 0.87 1-Methyl doublet 8.86 0.13 sec-C4H3D0H l-Methyl total 8.87 0.99 4-Methyl total 9.10 1.00 CH3CH2CH(CH3)0H 1-Methyl total 8.87 0.99 4-Methy1 total 9.10 1.00 56 compete with solvent capture of either 2-butyl cation. CH3CH2CDCH3 e;://// NH2 '\\‘\\33 + + CH3CH2CHCH2D . -4= CH3CHCH2CH2D 10 i J ( ) OH ?H I CH3CH2CHCH2D CH3CHCH2CHD As there is no deuterium in the product alcohols other than at C-2 and C-3, the composition of the derived 2-butanol can be calculated from the (P-C2H5) ion. HHZ 0H 6H CH3 CH2 CDCH3 __'9i CH3 CH2 CDCH3 + CH3 CHCI‘IDCHa 2.2. 2,3. 100% 88% 12% Although less accurate, nmr data are in essential agreement. By comparing the integrated signal of the trip- let (J : 0.9 cps) due to the 1-methyl protons of gg'to the total signal of 1-methyl protons, a concentration of 87 i 3% is calculated. As Figure 9 illustrates, resolution is not complete between the triplet and the doublet (J = 6.2 cps), due to the 1-methyl group of 33; 2-Butyl-3,3-g3 System. The expected similarities of this model and the 2-butyl-25g_system were verified exper- imentally. The ratios of protons at C-1 to those at C-4 were identical (Table XXIII) for derived 2-butanol and 2- butanol-go. That no deuterium migration to C-1 (of the 57 §_ CHSCHZCDCHS 05i(CH3)3 + gg_ CHSCHDCHCH3 m A 051(CH3)3 —> H0 .87 a m- Figure 9. 1-Methyl region of the nmr spectrum of the tri- methylsilyl ether of the 2-butanol obtained from 2-butyl-Zegfamine. 58 Table XXIII. Integrated nmr signals of various protons of the 2-butanol obtained from the deamination of 2—butyl-3,31d2-ammonium perchlorate and of unlabeled 2-butanol. Signal . Compounds Protons T Intensity Ratios sec-C4H7D20H 1-Methyl triplet 8.87 0.8 0.11 1-Methy1 doublet 8.86 6.2 0.89 sec-C4H7D20H l-Methyl total 8.86 7.0 1.03 4-Methyl total 9.10 6.8 1.00 CH3CH2CH(CH3)OH 1-Methyl total 8.86 7.0 1.03 4-Methyl total 9.10 6.8 1.00 59 alcohol) occurred was again indicated by the lOW'ép contri- bution of the (P-CH3) ion [0.9% Q”; (P-CH3) ion of authentic 2-butyl-3,3-§2 trimethylsilyl ether was also 0.9%‘d9]. Therefore, the composition of alcohols is calculated from the (P-C2H5) ion to be the following: NH2 on pH CH3CDZCHCH3 > CH3CD3CHCH3 + CH3CDCHDCH3 2:1. as. 100% 90% 10% Resolution of the nmr l-methyl signals of gé'and 32' was again incomplete (Figure 10); however, concentration of gg'in the 2-butanol mixture is estimated by integration to be 89 i 3%, in accord with mass spectral calculations. The above observations concerning the 2-butyl-syStems do not exclude the possibility of formation of the bridged- methyl ion 36 (see eq. 11 regarding the 2-butyl-2-g_system). CH3CH2CDOHCH3 + CH3CHDCHOHCH3 F x .— QZ. f ii CH3CH2CDCH3 CH3 CH2“; I or —> I, +\\\ 12> /\ \\ /CH \\ 2+ID + CH2;:_:C'ID CHz—CD CH2——CH CH CH CDCH L. 3 2 =2 g2 en. {,1 2m. 2m, 39' H (n) + CHZDCHZCHCHa [I’CH2 /CH2 + 7H3 H87 \ or /+ \\ + CHZ-CD <—— ‘CH2———CD CH2--CHD —> CHZCHZCHDCHa H3 CH3 CH3 60 ,A 2.. A. CH3CD2CHCH3 OH + El. Al CH3CHDCDCH3 OH Figure 10 . I T 8.87 T 9.10 Methyl region of the nmr spectrum of the 2-butanol obtained from 2-butyl—3,37Q2-amine. 61 Indeed, observed (28,48) formation of methylcyclopropane strongly implicates a bridged-ion species of some kind. The absence of 1-butanol products casts doubt on involve- ment of protonated cyclopropane intermediates, such as Q1, but the evidence is not prohibitive. Equilibration of 81' with other hydrogen- or alkyl-bridged intermediates can be considered unimportant, however, as 1-butanol, isobutyl alcohol and 2-butanol with deuterated methyl groups were not detected. Any 2-butanol produced from 82 would not be distinguishable from 82” formed by simple solvent capture of the original 2-butyl cation or displacement on its precursor, or from ggfi formed by capture after a 1,2-hydride shift. It has been reported (51,52) that a displacement mechanism accounts for SE: 22% of the 2-butanol-2ag.formed from Z-butyl-Z-g-amine. Results of the 2-butyl-3,3-§_2 sys- tem are consistent. That a 1,2-shift competes less suc- cessfully with capture than in the 2-butyl-2jg system can be ascribed to an isotope effect that is discussed later. The data gathered from the deaminations of deuterated 1- and 2-butylamines correlate with those concerning the deamination of isobutylamine. It was demonstrated (53) that the alcohols 82 and 85 were the only 2-butanols formed from isobutyl—1,1-g2—amine. The isobutyl alcohol formed was isotopically unrearranged. Thus, involvement of hydrogen — bridged intermediates was not substantiated, although g8. may be a possible representation for the precursor of methyl— cyc10pr0pane (28), and perhaps for some of the alcohol Q2, as well. 62 r . 9‘53 \ CH3CHCD2X ’///————-——> CHacH;—‘-'—'=CD2 38 or W l? ?H3 \\ + + 1,2— ~ — ~ CH3CHQD2 J M8 > CH3CHCD2CH3 ‘%*gJQ—> CH3CHDCDCH3 h + b , l 4 cns oH on CH3CHCD2NH2 > CHSCHCDZCH3 + CH3CHDCDCH3 22, ~22 100% 92% 8% Similar results have been obtained from'deamination of isobutyl-Zidfamine (54). In this case, also, the 2-butanol was formed solely by way of a Z-butyl cation which was the product of a 1,2-methyl shift. The data did not support the intermediacy of protonated cyclopropanes. CH3 CH3 /’+\\ I \ CH3CDCH2X > CH3CD====CH2 / 39 l? CH3CD$H2 \——-> CH3CDCH2CH3 '———————> CH3CHDCHCH3 h < _J t _ 1 1 $H3 OH ?H . . CH3CDCH2NH2'—---> CH3CDCH2CH3 CH3CHDCHCH3 §E E2 100% 89% 11% 63 Any correlation of the results of this investigation with those of previous workers must reconcile the formation of hydrocarbons (7,42,43) with that of substitution products. As was noted earlier, intermediate protonated cyc10propanes (possibly edged-protonated, and equilbrating through methyl- bridged ions) satisfactorily accommodated the formation of cyclopropane and isotope-position rearranged 1-propanol (20,21,22). On the other hand, the ion 42 has not been detected, if present, and 41 has been ruled out as an inter- i mediate in the deamination of neopentylamine (39). Iona c112.“ I, + \\ / \\+I’ (CH3 )2C_-.=_-.=CH3 (CH3 )2C-——-CH2 22. 2.1. The butyl systems exhibit properties in deamination reactions that are intermediate between those of the propyl and the neopentyl systems. Methylcyclopropane has been characterized as a deamination product of 2-butylamine, iso- butylamine and 1-butylamine in yields of 4%, 2.5% and 0.3% (28) of the hydrocarbon fraction, which was reported to be .33. 12% of the total product mixture in the 50% aqueous acetic acid deamination (55). Furthermore, isotopic analyses of methylcyclopropane obtained from the deamination of deuterated 1-butylamines in aqueous acetic acid have indi- cated (56) that a methyl-bridged ion (as 42 below) cannot be a sole precursor of the methylcyc10propane. Thus, the 1-butyl-1,1-§2-amine gave 9% gf-and 91% gz-methylcyclopropane, 64 while values of 11% £1 and 89% d2 were obtained from the 1-butyl-2,25g2 system. These data give credence to a mechanism involving equilibrating protonated cyclopropanes without, necessarily, the intervention of a methyl-bridged species (sequence 12). Such an equilibration would lead to -92 ‘L cnacnzcnzcnzx T-H+ 1 \\\s $33 9H3 on ‘~ cgi‘. + CH3 \ +IIH _.:_. / \+ /D .9...) -d1 / CH2——\Cf)2 CH2—-\CfID - Icgz ‘f// I + \\ CH2"‘CDz 1T (12) t c\ - / \\+:’H L d cmacn2 CD2CH2X —-> c132——-CH2 -2 equal amounts of methylcyclopropanefgl from 1—butyl-1,1-d2- amine and 1-butyl-2,2-g3-amine. The 9:1 ratio of gz- to ‘gl-species might reflect not only the relative ease of H+ to D+ loss resulting from a statistical factor (4 H to 2 D) and an isotOpe effect after equilibration, but also competi- tion between methylcyclopropane formation and species equi- libration. This competition factor would explain the 65 results of 43% g4- and 57% gz-methylcyclopropane afforded by deamination of 1-butyl-3,35§2-amine. Thus, the ion 42' would be the most important precursor of the cyclic product. (EH3 CH3 I / \\ +‘ ‘D ‘ \ +“H CH3CD2CH2CH2X > \ /’ < > \\ // 2 etc. CH2__CH2 CHz—CHD 2:: 1 l <91 ‘92 A similar phenomenon has been observed in the deamination of 1-pr0pylamine (22) and in the formolysis of 1-propyl tosylate (57), in which cases solvent capture competes with the equilibration of the protonated cyc10propanes. Such equilibration before H+ (or D+) loss to give methylcyclo- propane can account for the small difference detected (48) in deuterium distribution between methylcyclopropane and 1-butene in the aprotic deamination of isobutyl-1,1-g3- amine. In this instance, the olefin, which could not be formed with correSponding loss of deuterium, always contained slightly more deuterium than the methylcyclopropane, which could be produced with loss of deuterium by means of a series of equilibrations as depicted in eq. 12. It was also disclosed (56) that 1-butyldbL4-93-amine gave only trideuterated methylcyclopropane. This finding indicates that a 2-butyl cation may not be the main pre- cursor of the bridged species. If the bridged species were 66 3% one + / \ .2)!» cnscnzcrizcnzx > CD3CH2CHCH3 —H—> \ , 1 CH2—"—'CD2 + l-H 3-D+ CD3 $H3 AB //CE\ CH2— CH2 CH2 CD2 formed at least 100 times faster from the cationic pre- cursors (g;g,, diazonium ions) than from the cations, then the much greater amounts of cyclic products produced from 2vbutylamine (28) and 3—methyl-2-butylamine (26) than from l-butylamine and isopentylamine, respectively, would be reasonable. An SNi-like mechanism, with the leaving group being internally displaced by (a) the bonding electrons of the y—carbon and one of its hydrogens to lead to a protonated cyclopropane, or (b) by the electrons comprising the bond between CB and C to result in a carbon-bridged Species, would accommodate the results. One might consider (a) and (b) to be pathways competing with each other as well as with / \ / 1‘g ‘~°c’*H ‘t/ I \ Y \~“~~ . £1121. \ /b\\a 121.. \ / ,H —’ ‘rc-———c “ ’ >c——'C' """" \ I B < /H§\ \CI {QNz simple dissociation of the diazonium ion. Dissociation also would compete with concurrent migration or elimination of a function on C6. 67 By using the 1-propyl system (no substitution on Ca, C , or Cy) as a base, the protonated cyclopropane pathways 6 for isobutyl (methyl on C ) and 1-butyl (methyl on Cy) 5 would be less significant than the corresponding one for 1-propyl because of 1,2-eclipsed interactions between hydro— gen and methyl in the latter systems. In the neopentyl system, the carbon-bridge route would be favored over the hydrogen-bridge route because the tertiary CB can suc— cessfully stabilize the incipient positive charge. Indeed, it may do this so well that the bridged species may be merely a transition state in the conversion to a tramyl cation (39). By taking into account all of the available data, a scheme can be formulated which accommodates the results obtained from the deamination of 1-butylamine. It is illustrated in Figure 11. The dotted arrows represent pathways for which direct experimental evidence has not been obtained. A mechanistic implication of Figure 11 is that the alcohols are not only irreversibly formed, but also that they are formed from unrearranged or rearranged carbonium ions that have precursors derived from the starting amines, not from other reaction products. That hydration of the olefins in not such a pathway will now be demonstrated. The mass spectral data from the 1-butyl-2,2—§2 system permit a check on the extent of sequence 13. The (P-C2H5) ion of gé'would be a Q; species, whereas those of gg'and 32' 68 C4H90S + C4H8 products HOSI -H+‘ [1.2-H~ + os os CH3CH2€H2CH2 CH3CH2CHCH3 CH3CHCH2CH3 A\ I I _ x I Hos THOS I 1, 2-H + CH3CH2CH2CH2X -————> CH3CH2CHCH3 : —> CH3CHCH2CH3 w" A A Q. I I I CH3CH2CH=CH2 CH3CH=CHCH3 CH2=CHCH2CH3 I I ‘W CH3 cHa cH3 cHa I [CH2 cH~ : ‘ +CH / +‘\ f::; // ‘ *' H “‘* //C‘ «\+':Hb ---> // CH2===CH2 CHZ—CH-HB CH2—-—CH-Ha CH2——CH3 / \ / \H03 1 HOS/ \Ha I: Hb \ HOS / N u os CH3CH2CH2CHZOS Afib AH CH3CHCH2CH3 a (unrearranged) + CH3CH2CH2CHZOS (rearranged) Figure 11. Proposed mechanisms of formation of the products of the deamination of l-butylamine. 69 CH3CH2CD20H2X OH OH I I or > CH3CH2CDCH2D + CH3CHCHDCH2D + CH3CH2CD2CH2 gg' g2, "‘ (13) I E2 or E1 . H+ + + CH3CH3CD=CH2 > CH3CH2CDCH3 < > CH3CHCHDCH3 SE. I I (3H (3H CH3CH2CDCH3 CH3CHCHDCH3 :12, 1:», would be 93 and g”, respectively. Table XIX shows that the g1 (P-C2H5) ion contribution of the product 2—butanol from 1—butyl-2,2—d2-amine (which contained 3.7% amine) was 3.9%. Thus, alcohol 44 is not a significant product. It can be likewise established that 2-butene is not a 2-butanol precursor (sequence 14). Alcohols 41 and 48 would produce — , CH3CD2CH2CH2X OH OH I or > CH3CD2éHCH3 'I' CH3CDCHDCH3 + CH3CD2CH2CH2.J gg, g1 ‘ (14) I1,2-H~ + CH3CD2CHCH3 + I'D + pH on CH3CD=CHCH3 H > > CH3CDCH2CH3 + CH3CHDCHCH3 12. 1:2. 70 a d; (P-CH3) ion. As the experimental value of 3.7% d1 is less than the 4.3% present in the starting material, sequence 14 is excluded. The unimportance of olefin hydration was not unexpected, as Karabatsos, Orzech and Meyerson (39) had shown that no .E-amyl products arose by such a pathway in the deaminative rearrangement of neopentylamine. Also, Otvos and coworkers (58) had shown that isobutylene was converted only very slowly into carbonium ions in 96% sulfuric acid. This investigation was not designed to yield specific information on the mechanism of deamination prior to the dissociation step, or even of the dissociation itself. However, the ratios of the various isotopically-distributed 2-butanols obtained in the deaminations of the 1-butyl, 2-butyl and isobutyl systems were found to be curiously different, and may be indicative of the characteristics of the dissociation and the steps immediately following it (42,59). In Figure 12 is shown the 2-butanol formed from seven related butyl models. It should be noted that because the "second" 2-butyl cation formed can rearrange in competition 'with being captured by solvent, the analyzed ratios of alcohols are typical. but not exact, measures of the extent of the second 1,2-shift, relative to capture of the "first" 2-butyl cation by water. Many of the differences in the ratios are readily interpreted as due to deuterium isotope effects. Thus IIIb 71 Rela- Case 2-Butanols No. tive % OH 1,2-H~' + _——_> I I CH3CH2CH2CD2X CH3CH2CHCHD2 —> CH3CH2CHCHD2 Ia 77 II ?H + CH3CHCH2CHD2 —> CH3CHCH2CHD2 lb 23 OH 1,2-D“’ + I II CH3CH2CD2CH2X ———————» CH3CH2CDCH2D -> CH3CH2CDCH2D 11a 75 II OH + I CH3CHCHDCH2D —> CH3 HCHDCHZD IIb 25 OH 1,2-H~' + I III CH3CD2CH2CH2X‘-—-——-> CH3CD2CHCH3 -> CH3CD2CHCH3 IIIa 81 II on + I CH3CDCHDCH3 -> CH3CD2CHCH3 IIIb 19 . <.>H Iv CH3CH2cDCH3 ----> CH3CH2CDCH3 -> CH3CH2CDCH3 IVa 88 x + I I cum CH3CHCHDCH3 -> CH3CHCHDCH3 IVb 12 . 33H V CH3CD2cHCH3 -———————> CH3CD2CHCH3 -—> CH3CD2CHCH3 Va 90 X + II OH I CH3CDCHDCH3 -> CH3CDCHDCH3 Vb 10 OH 1,2-Meea + I VI CHacHCDZX ———————¢ CH3CD2CHCH3 —> CH3CD2CHCH3 VIa 92 CH3 + II OH ‘ I CH3CDCHDCH3 -> CH3CDCHDCH3 VIb 8 OH 1,2-Menv + I VII CH3?DCH2X '—————-> CH3CH2CDCH3 —> CH3CH2CDCH3 VIIa 89 CH3 II 0H + I CH3CHCHDCH3 —> CH3CHCHDCH3 VIIb 11 Figure 12. Ratios of labeled 2-butanols obtained from the de- aminations of deuterated 1-butylamine, 2-butylamine and iso- butylamine. 72 is expected to be formed to a lesser extent than Ib or IIb since the second 1,2-shift involves deuterium rather than protium and, consequently,would compete less successfully with solvent capture. For the same reason, Vb is less than IVb and VIb less than VIIb. Furthermore, IIb may have been detected in a slightly greater amount than Ib because of the less likelihood of the cation precursor of IIb under- going a third 1,2-shift than that of Ib. The finding that the ratios VIb/VIa and Vb/Va are smaller than the correSponding ratios IIb/IIa and IIIb/IIIa can be partially explained by direct diSplacement on the diazonium ion of IV and V, to lead to larger amounts of IVa and Va. Yet assuming a value of 22% (51) for the amount of 2—butanol arising by direct displacement (leading to IVa, for example, in case IV) the percent of IVb should have been 20-22%, not 12% as found. There are other differences in the values determined which are also less easily rational- ized. For example, apparently similar 2-butyl cations are initially formed in cases III and VI (by a 1,2—hydride shift in case III and a 1,2-methyl shift in case VI). Yet a second 1,2-shift was more than twice as competitive in case III than in case VI. The same is true in cases II and VII. The aforementioned differences can be explained as follows. The low value obtained for IVb is most easily rationalized in terms of competing direct displacement on the diazonium ion and carbonium ion formation. Conceivably 73 the SNZ pathway accounts for more than the 22% reported (51). A value of 40-45% displacement as a pathway to 2-butanol in the 2—butyl systems would make the 12% IVb comparable to 25% 11b. The low amount of VIIb formed in the isobutyl-ng case can be accounted for in terms of the stereochemistry of the first 2—butyl cation. The intially-formed 2-butyl cation (42) from the l-butyl system can assume a geometry where a second 1,2-shift is favorable. A slightly dif- ferent situation pertains to the isobutyl system. The D \é—DC/l/H 1 2-D~ D/; c c D H D” \ / Q + 1" C/+ :\H C 2 I H ///I\\ //C \\.H N2 CH3 H CH3 \1 2-H~ / 22. H D In \ / CH3CH2CDCH2D D--—C_C\ + \ IIa(75%) /C< \H H CH3 H I (')H CH3CHCHDCH2D IIb(25%) 2-butyl cation (52) formed by the 1,2-methyl shift is not of appropriate conformation to facilitate a subsequent 1,2— hydride shift. Consequently, the required rotation about 74 CH3 p CH3 IIIH 1 L2-Me~ A + / /C—C 7 ’zc——C\“H ’ I mix I EN + CHSD/ E H 3 2 E D 2 / 22. 1 2-..- 9H CH3CDCH2CH3 H\\ + c-———c”’H “UCH3 VIIa (89%) ‘3? CH5’ D / oH CH3CHDCHCH3 VIIb (11%) the middle carbon-carbon bond may make the second 1,2-shift less competitive with solvent capture of the initially- formed 2-butyl cation (Q2). V. Summary The following points were brought out during the course of this investigation. i. All of the 1-butanol produced in the deamination of 1-butylamine arises from displacement by solvent on a 1-butyl cation precursor (such as a diazonium ion) and/or from capture by solvent of an unrearranged 1-butyl cation. V ii. The mechanism by which 2-butanol is formed from 1- butylamine involves only 1,2-hydride shifts, with no signifi- cant contribution of a nominal 1,3-hydride shift. 75 iii. The only alcohol formed in the deamination of 2-butyl- amine is 2-butanol, which results from diSplacement by solvent on a 2—butyl cation precursor and from solvent capture of interconverting Z-butyl cations. iv. The carbonium ions leading to the alcohols have precursors derived from the starting amines, not from ole— fins. Diazoalkanes or carbenes are not significant inter- L— mediates in the alcohol formation. EXPERIMENTAL I. General Analytical vapor phase chromatography (v.p.c.) was performed on an F & M Model 700 or on an Aerograph Model A-90-P gas chromatograph. The latter instrument was also used for small-scale preparative v.p.c. Larger scale preparative work was done on a Perkin-Elmer Model 154 Vapor Fractometer. Nuclear magnetic resonance spectra were taken on a Varian Model A-60, high resolution spectrometer. II. Preparation of 1-Butyl-1,1fQ2-ammonium Perchlorate To a stirred slurry of 4.0 g (0.095 mole) of lithium aluminum deuteride (Metal Hydrides, Inc.) in dry ether was slowly added an ethereal solution containing 6.6 g (0.096 mole) of butyronitrile. Stirring was continued at reflux for four hr after addition was complete. The ether phase obtained by filtration,after alkaline treatment of the re- action mixture,was combined with ether washings of the in- organic salts and neutralized with 71% perchloric acid. Removal of solvent at reduced pressure left white crystals which were washed with pentane and dried in vacuo, afford- ing 15.2 g (yield, 90%) of 1-buty1-I,1-g2-ammonium per- chlorate, mp 186-1880. 76 77 III. Preparation of 1-Butyl-2,2—g2—ammonium Perchlorate A. Preparation of Butyric-2,2-g2 acidjd To 1180 g of absolute ethanol in which 69 g (3.0 g—atoms) of sodium had been dissolved was added 480 g (3.0 moles) of diethyl malonate, followed by 343 g (3.15 moles) of bromoethane. The resulting mixture was neutral after 0.5 hr reflux. Rotary evaporation of the filtered alcoholic solution left a two-phase residue of sodium bromide, ethanol and diethyl ethylmalonate, which was extracted with ether. The ether extracts were combined and the solvent removed. Unreacted diethyl malonate was removed by shaking the pro— duct ester for one minute with 50 ml of cold, 20% sodium hydroxide solution. The yield of diethyl ethylmalonate was 535 g (95%). The diester was hydrolyzed by adding it to a hot solu- tion of 525 g of potassium hydroxide in 500 ml of water. Ethanol was removed continuously by a vacuum line. The mixture resulting from neutralization with concentrated hydrochloric acid was extracted several times with ether and ethyl acetate. Removal of solvent left 309 g of ethyl- malonic acid, which was exchanged six times with deuterium oxide. Nmr indicated the absence of a methine proton. A portion of the ethylmalonic-Z-g_acid—§2 was decarboxylated by distillation at atmospheric pressure. About 41 g of butyric-2,2—g2 acid-d_was obtained at 150-1570. 78 B. Preparation of Butyronitrile-2,2-§b To 36.0 g (0.40 mole) of butyric-2,2—§2 acidfig_was added 50.7 g (0.43 mole) of thionyl chloride that had been purified by successive distillations from quinoline and linseed oil. After two hour reflux, the mixture was taken up in 500 ml dry ether and cooled. Ammonia was passed over the ethereal solution fo 1.5 hr. The butyramide-2,2-gb was removed from the inorganic salts by Soxhlet extraction with benzene. After recrystallization from benzene, 24.0 g (yield, 68%) of product was obtaind. Dehydration of 20.3 g (0.23 mole) of amide to butyro- nitrile-2,2-d2 was accomplished by refluxing it with 40.7 g (0.34 mole) of thionyl chloride at 1000 for 3.5 hr. Addi- tion of water (to hydrolyze excess thionyl chloride) to the mixture which was taken up in ether was followed by washing of the ether phase with saturated sodium bicarbonate solution and drying over anhydrous magnesium sulfate. After distillation, the yield was 11.2 g (70% from amide) of butyronitrile-2,2-d2. C. Preparation of 1-Butyl-2,2-g3-ammonium Perchlorate An ethereal solution containing 11.1 g (0.156 mole) of butyronitrile-2,2-g2 was slowly added to 6.1 g (0.16 mole) of lithium aluminum hydride in dry ether. The mixture was refluxed 3 hr and hydrolyzed‘with alkali. The ether layer was decanted, added to the ether washings of the inorganic 79 salts and neutralized with 71% perchloric acid. Removal of solvent at reduced pressure gave white crystals, which were washed with petroleum ether and dried in vacuo to give 17.9 g (yield, 65%) of 1-butyl-2,2:§2-ammonium per- chlorate, mp 192-1930. IV. Preparation of 1-Butyl-3,3-g2-ammonium Perchlorate A. Preparation of Propionic-2,2-g2 Acid-d Method A. Bromomethane, generated by warming a mixture of 770 g of concentrated sulfuric acid, 908 g of sodium bromide and 600 g of methanol, was bubbled for 3.5 hr into a mixture of 1180 g absolute ethanol, 69 g (3.0 g-atoms) of sodium and 480 g (3.0 moles) 6f diethyl malonate. After an additional two-hour reflux period, ethanol was removed by rotary evaporation and the diethyl methylmalonate was extracted from the aqueous solution of the inorganic salts with ether. Removal of solvent, hydrolysis with hot potas- sium hydroxide solution and acidification with concentrated hydrochloric acid gave 251 9 (yield, 71%) of methylmalonic acid. The product of six exchanges with deuterium oxide was decarboxylated by refluxing for 12 hr at 135°. Distil- lation (130-1400) gave 100 g of propionic—2,2-g2 acidfd (yield, 43% from diethyl malonate). Method B. Iodomethane (500 g, 3.52 moles) was added in a two-hour period to a mixture consisting of 79 g (3.44 g- atoms) of sodium dissolved in 1400 g of absolute ethanol. 80 After a 0.5 hr reflux period, most of the ethanol was re- moved by rotary evaporation. Extraction with ether, which was subsequently removed, yielded 610 g of wet diethyl methylmalonate, which included small amounts of unmethylated and dimethylated diesters as well. Basic hydrolysis, neu— tralization with hydrochloric acid, and exchange of the resulting diacid with deuterium oxide gave methylmalonic-2- §_acid:2, which was decarboxylated by refluxing at 140° for 50 hr. Distillation gave 164 g (yield, 62% from diethyl malonate) of propionic-2,2-§2 acidfid. B. Preparation of 1-Propanol-2,2-g2 #— To a slurry of 61 g (1.6 moles) of lithium aluminum hydride in dry ether was added 96 g (1.25 moles) of propionic- 2,2-d2 acidfid. .Alkaline hydrolysis of the mixture stirred at room temperature for 8 hr gave an ether phase which was decanted, dried over anhydrous sodium sulfate and distilled to give 49 g (yield, 85%) of 1-propanol-2,21d2. C. Preparation of 1-Bromopropane-2,2-g2 Method A. To 39.6 g (0.64 mole) of 1-propanol-2,2-gb was added 75.0 g (0.27 mole) of phosphorus tribromide at 0°. The mixture was stirred at room temperature for 18 hr. After adding 60 ml of water, the top (aqueous) layer was extracted with ether and the combined extracts were added to the lower layer. The ether phase was dried and distilled, giving 24 g (yield, 30%) of 1-brom0pr0pane-2,2—g2. 81 Method B. Twenty-seven grams (0.10 mole, 13% excess) of phosphorus tribromide was added in 50 min. to ice- cooled 1-propanol-2,2-g2 (16.4 g, 0.264 mole). After stir- ring at room temperature for 22 hr, then at gentle reflux for 1 hr, 30 ml of water was added to the two-phase liquid mixture. The clear, lower (bromide) layer was separated and washed with water and saturated solutions of sodium chloride and sodium bicarbonate. The 1—bromopropane-2,25g2 (26 g, 80% yield) was purified by distillation at 68-69°. D. Preparation of Butyronitrile-3,35Q2 A solution of 20.0 g (0.16 mole) of 1—brom0propane- 2,2-g2 in 26 ml of methanol was added to 8.6 g (0.18 mole) of sodium cyanide dissolved in 11 ml of water. After re- fluxing 20.5 hr at 115°, the mixture was extracted continu— ously with ether for 40 hr. The ether phase was washed with solutions of hydrochloric acid, sodium bicarbonate and sodium chloride (which removed side-product isonitrile as well as methanol), dried and distilled, yielding 10 g (80%) of butyronitrile-3,3-g2. E. Preparation of 1-Butyl-3,3-g2—ammonium Perchlorate Butyronitrile-3,3-g2 (5.41 g, 0.076 mole) in ether was slowly added to 3.6 g (0.095 mole) of lithium aluminum hydride in dry ether. The reaction mixture was stirred at reflux temperature for 4.5 hr, then at room temperature for 4 hr, and hydrolyzed with base. After standing overnight, 82 the ether phase and ether washings of the aqueous paste were combined and neutralized with 71% perchloric acid. White crystals of 1-butyl-3,3fig2-ammonium perchlorate, mp 189-191°, resulted from solvent removal and washing with petroleum ether. The yield was 11.8 g (89%). V. Preparation of 2-Butyl-2-gfammonium Perchlorate To a slurry of 4.9 g (0.12 mole) of lithium aluminum deuteride in dry di-nfbutyl ether was added 10.2 g (0.12 mole) of 2-butanone oxime in di-gybutyl ether. After stir- ring for,5 hr at 150°, the brown product mixture was hydro- lyzed-with base. The decanted ethereal phase was dried and distilled to give a 4.5 g fraction, bp 65-700, containing 85% 2-butyl-2-g-—amine (yield,50%). Neutralization with 3.4 ml 71% perchloric acid (0.061 mole) gave a slurry of perchlorate salt in di-nfbutyl ether. Because removal of solvent was difficult, the salt was dis- solved in water and the organic materials were removed by extraction with ether. The resulting aqueous solution was then subjected to deamination conditions. VI. Preparation of 2-Butyl-3,3-g3-ammonium Perchlorate A. Preparation of Ethanol-1,1-g2 *— An ethereal solution of 30.4 g (0.298 mole) of acetic anhydride was slowly added to an ice-cooled slurry of 11.4 g (0.271 mole) of lithium aluminum deuteride in dry ether. 83 The mixture was refluxed for 8 hr, hydrolyzed with base, and stirred overnight. The ether phase was decanted and combined with the Soxhlet ether extracts of the inorganic salts. Distillation, after drying over anhydrous magnesium sulfate, gave 15.9 g (yield, 62%) of ethanol-1,15g2. B. Preparation of Bromoethane-1,1-§2 t:— Phosphorus tribromide (34.9 g, 0.129 mole, 16% excess) was added dropwise to 15.9 g (0.332 mole) of ethanol-1,1-d2 at 0°. The mixture was brought to room temperature and stirred for 45 hr. After refluxing gently for 45 min. and adding 35 ml of water, the mixture was transferred to a separatory funnel and the lower bromide layer drawn off. The yield of bromoethane-1,1-d2 was 22.2 g (60%). C. Preparation of 1-Butanol—3,35g2 *— To 24 g (0.22 mole) of magnesium turnings was added 22.2 g (0.20 mole) of bromoethane-1,1-g3 in dry ether. The resulting grey solution was refluxed for 0.5 hr and cooled. To it was added 9.70 g (0.20 mole) of acetaldehyde in ether. After refluxing for 1.5 hr and hydrolyzing with aqueous am- monium chloride, the mixture was separated into an ether phase and a slurry of inorganic salts. The salts were ex- tracted thoroughly with ether. The ether phases were com- bined, dried, and distilled to give 8.2 g (yield, 54%) of 2-butanol—3,3—d2. 84 D. Preparation of 2-Butyl-3,3—g2pftoluenesulfonate A mixture of 23.0 g (0.121 mole) of pftoluenesulfonyl chloride and 40 ml of pyridine was added slowly to 8.2 g (0.12 mole) of 1-butanol-3,3-g3 in 20 ml of pyridine. The reaction mixture was maintained at -5°, or below, during the addition and subsequent 0.5 hr period of stirring. After allowing to stand at -100 for 17 hr, the mixture was added to 50 ml of water and the organic materials were ex- tracted with ether. Careful washing of the combined ether phases with aqueous solutions of hydrochloric acid (6 fl) and sodium chloride was followed by removal of solvent by rotary evaporation. About 24.8 g of 2-butyl-3,3-g2 pf toluenesulfonate (yield, 90%) was obtained. E. Preparation of 2-Butyl-3,3-§2-azide A mixture of 11.0 g (0.169 mole) of sodium azide, 40 ml of water, and 100 ml of methanol was heated to 60°. To it was added 14.8 g (0.108 mole) of 2-butyl-3,3-§2 pf toluenesulfonate in 40 ml of methanol. Reflux was maintained during the addition and for another 16 hr. After cooling the product mixture to room temperature, 80 ml of water was added and the resulting mixture was taken up in ether. A solution consisting of 40 g of calcium chloride in 100 ml of water was added and the resulting phases were separated. The ether layer was combined with the ether extracts of the aqueous phase and dried over anhydrous magnesium sulfate. Because of its unstable properties the azide was not isolated. 85 F. Preparation of 2—Butyl-3,3:d2-ammonium PerChlorate The ethereal azide described above was added to a cold slurry of 5.0 g (0.13 mole) of lithium aluminum hydride in dry ether. After refluxing for 4 hr,tfle mixture was hydro- lyzed with base and stirred overnight. The ether layer was decanted and combined with the ether washings of the inor- r ganic salts. Neutralization with 71% perchloric acid and * evaporation of solvent left a brown slurry containing ca. 3.1 g of 2-butyl-3,3-§2-ammonium perChlorate (yield, 30% ( from 2-butyl-3,3-g3 pftoluenesulfonate). The perchlorate salt was not isolated and purified, but was dissolved in water and the solution was extracted thoroughly with ether. After removing remaining traces of ether by heating on a steam bath for 5 min, the aqueous solution was subjected to deamination conditions. VII. Preparation of Authentic, Deuterated Alcohols. A. Preparation of 1-Butanol-1,lfig£ An ethereal solution containing 1.1 g (0.012 mole) of butyric acid was slowly added to a cold slurry of 0.41 g (0.0098 mole) of lithium aluminum deuteride in dry ether. After addition, the mixture was stirred at reflux for 4 hr and allowed to stand overnight. After alkaline hydrolysis the ether phase was dried over anhydrous magnesium sulfate and distilled to give 0.2 g of 1-butanol-1,1-d3. 86 B. Preparation of 1-Butanol-2,2-g2 To cold slurry of 1.35 g (0.036 mole) of lithium alum- inum hydride in dry ether was slowly added 4.3 g (0.047 mole) of butyric-2,2-d2 acid-d, After 12 hr of reflux and alka— line hydrolysis, the ether phase was decanted, dried, and distilled, affording 0.25 g of 1-butanol-2,2-§2. C. Preparation of 1-Butanol-3,3jd2 A mixture of 0.80 g (0.011 mole) of butyronitrile—3,3—g3, 5 ml of 85% phosphoric acid and 3 ml of 75% sulfuric acid was slowly heated to 165°, at which temperature it was re- fluxed for 1.5 hr. Extraction with ether of the mixture that was diluted with water afforded an ethereal solution of butyric«3,3-d2 acid, which was dried and added to 0.43 g (0.11 mole) of lithium aluminum hydride in dry ether. Careful distillation of the decanted ether phase yielded 0.14 g of 1-butanol-3,3-d2. D. Preparation of 1-Butanol-24d To 0.94 g (0.022 mole) of lithium aluminum deuteride in dry ether was added a solution of 1.61 g (0.022 mole) of l-butanone in ether. The mixture was refluxed for 6 hr, then allowed to stand overnight. After alkaline treatment, the ether phase was separated from the aqueous paste, dried and distilled. The yield of 2-butanol-2-g_was 0.9 g. 87 VIII. Deamination of Butylammonium Perchlorates A typical run consisted of placing 21.5 g (0.123 mole) of deuterated 1-butylammonium perchlorate, 15.4 g (0.109 mole) of 71% perchloric acid and 60 g of water in a 3-necked, 250 ml flask equipped with condenser, addition funnel, thermometer and magnetic stirrer. Sodium nitrite (17.7 g, 0.257 mole, Mallinckrodt) dissolved in 40 g of water was added dropwise over a two-hour period, during which the re— action temperature generally rose to 35-38°. After an additional three hours of stirring, the solution was salted out with sodium chloride and extracted with ether. The combined ether layers were washed with saturated solutions of sodium chloride and sodium bicarbonate. The ethereal solution, which usually remained straw yellow, was dried over anhydrous magnesium sulfate and distilled until a liquid residue of approximately 2 ml remained. This resi- due was fractionated on a preparative gas chromatograph, by using a nine-foot column of 25% Carbowax at 115°, which permitted collection of 1-butanol and 2-butanol either individually or together. If necessary, the alcohols were further purified by preparative gas chromatography at the above conditions. ‘ Hexamethyldisilazane (Metallomer Laboratories) and alcohol (either pure or a mixture of 1- and 2-butanol), in a 1:2 molar ratio, respectively, were refluxed overnight with a drop of chlorotrimethylsilane. The trimethylsilyl 88 ethers were then separated, if necessary, and purified by gas chromatography on a 20-foot Carbowax column at 85°. Deaminations of the 2-butylammonium perchlorates were carried out under the same conditions. Work up was also similar, the only difference being that the only alcohol detected (by analytical gas chromatography) was 2-butanol, which was collected and derivatized with hexamethyldisila- zane as above. 10. 11. 12. 13. 14. 15. 16. REFERENCES . Roberts and G. E. Kimball, J. Am. Chem. Soc., 55, 947 (1937); see also 8. Winstein and H. J. Lucas, ibid., 5%, 1576 (1939). ' . Winstein and H. J. Lucas, ibid., 52” 836 (1938). . P. Nevell, E. de Salas and C. L. Wilson, J. Chem. Soc., 1188 (1939). For general reviews of this topic see (a) B. Capon, J. Quart. Rev. (London), 18, 45 (1964); (b) G. D. Sargeant, ibid., 25, 36T’(1966); (c) M. J. s. Dewar and A. P. Marchand, Annual Review of Physical Chemistry, 15“ 321 (1965). . D. Roberts and M, Halmann, J. Am. Chem. Soc., 15, 5759 (1953). . D Roberts and J. A. Yancey, ibid., 24” 5943 (1952). C. Whitmore and D. P. Langlois, ibid., 54, 3441 (1932). D. Roberts and C. M. Regan, ibid., Z5” 2069 (1953). For a recent report on the existence of phenonium ions, see G. A. Olah and C. U. Pittman, Jr., ipid., 51, 3509 (1965). S. Skell and I. Starer, ibid., 52” 2971 (1960). S. Skell and I. Starer, ibid., 54, 3962 (1962). . A. Reutov and T. N. Shatkina, Dokl. Akad. Nauk SSSR. 133, 606 (1960); Tetrahedron, 15; 237 (1962). . J. Karabatsos and C. E. Orzech, Jr., J. Am. Chem. Soc., §g, 2838 (1962). . A. Reutov in "Congress Lectures, XIXth International Congress of Pure and Applied Chemistry", Butterworth and Co., Ltd., London, 1963, p. 303, cited in (19). . L. Baird and A. Aboderin, Tetrahedron Letters, 235 (1963). . L. Baird and A. Aboderin, J. Am. Chem. Soc., 55” 252 (1964). 89 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. F. 90 C. Whitmore and R. S. Thorpe, ibid., 22/ 1118 (1941). A. A. Aboderin and R. L. Baird, ibid., 55, 2300 (1964). C. C4 :3 3 t! 22 m o O C. Lee, J. E. Kruger and E. W. C. Wong, ibid., 51” 3985 (1965). . Lee and J. E. Kruger, ibid., 51” 3986 (1965). . Lee and J. E. Kruger, Tetrahedron, 25” 2539 (1967). . Karabatsos, C. E. Orzech, Jr., and S. Meyerson, . Am. Chem. Soc., 51” 4394 (1965). LJC-I O O . Hart and R. H. Schlosberg, $225,, 55” 5030 (1966) . C. Deno and D. N. Lincoln, ig;g., 55“ 5357 (1966). . J. Cram and J. E. McCarty, 325g;, 12, 2866 (1957). . S. Silver, ibid,, 52” 2971 (1960). . S. Silver, J. Org. Chem., 25“ 1686 (1963). H. Bayless, F. D. Mendicino and L. Friedman, J. Am. Chem. Soc. 87, 5790 (1965) . E. Edwards and M. Lesage, Can. J. Chem., ~5, 1592 (1963). . E. Edwards and M. Lesage, Chem. Ind. (London), 1107 (1960). . Winstein and H. Marshall, J. Am. Chem. Soc., 74, 1120 (1952). See also D. P. Stevenson, C. D.“Wagner, 0. Beeék and J. W. Otvos, ibid., 22” 3269 (1952). . D. Bartlett, Bull. Soc. Chim. FranCe, (5), 15” 0100 (1951). . D. Bartlett and M. S. Swain, J. Am. Chem. Soc., ZZ/ 2801 (1955). . Winstein and B. K. Morse, ibid. 74,1133 (1952). . Brown and R. B. Kornblum, ibid., 76, 4510 (1954); . c. Brown and Y. Okamoto, ibid. ,777'3619 (1955), . C. Brown and I. Moritani, ibid. ,ZZ', 3623 (1955). . Roberts and J. A. Yancey, ibid., 11, 5558 (1955). C1 U {1:530 . Karabatsos and J. D. Graham, ibid., 52” 5250 (1960). 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 91 M. 3. Silver, ibid., 99, 3482 (1961). G. J. Karabatsos, C. E. Orzech, Jr. and S. Meyerson, ibid., §§x 1994 (1964). P. S. Skell, I. Starer and A. P. Krapcho, ibid., Q2, 5257 (1960). P. S. Skell and I. Starer, ibid., 81” 4117 (1959). A. Streitweiser, Jr. and W. D. Schaeffer, ibid., 12” 2888 (1957). D. 0.6Rickter, Ph.D. Thesis, Michigan State University, 19 4. S. H. Langer, S. Connell and I. Wender, J. Org. Chem., 23” 50 (1958). L. P. Hammett, "Physical Organic Chemistry“, McGraw Hill Book Co., Inc., New York, N.Y., 1940, p. 295. E. D. Hughes, Bull. soc. chim. France, C17 (1951). Isobutanol formation from 1-butylamine has been dis- proven despite earlier claims by (a) P. c. Ray and J. N. Rakshit, J. Chem. Soc., 101, 141 (1912); (b) E. Linnemann and V. v. ZotEST’Ann,, 162“ 3 (1872). A. T. Jurewicz and L. Friedman, J. Am. chem. Soc., 82' 149 (1967). J. D. Roberts, R. E. McMahon and J. S. Hine, ibid., 22, 4237 (1950). The (P-CH3) ion includes a 21% contribution due to cleavage of methyl from the 2-butyl group, as dis- cussed earlier. K. B. Wiberg, Dissertation, Columbia University, July, 1950, cited in (42). P. Brewster, F. Hiron, E. D. Hughes, C. K. Ingold and P. A. D. S. Rao, Nature, 166, 179 (1950). G. J. Karabatsos, N. Hsi and S. Meyerson, J. Am. Chem. Soc., §§, 5649 (1966). G. J. Karabatsos, N. Hsi and S. Meyerson, Private Communication. 55. 56. 57. 58. 59. 92 Other workers have reported yields of total hydro- carbon products from 1-butylamine to be 16% (43) under conditions used during the present work and 36% (7) in hydrochloric acid. L. Friedman, private communication. C. C. Lee and J. E. Kruger, Can. J. Chem., 44, 2343 (1966). J. W. Otvos, D. P. Stevenson, C. D. Wagner and O.. Beeck, J. Am. Chem. Soc., 22/ 5741 (1951). A. Streitweiser, Jr., J. Org. Chem., 22” 861 (1957). MICHIGAN sure UNIVERSITY LIBRATIES lllllllmllWl WIW)||Will)llI)\IIWIIHIHHIHIH 31193 03142 9594