Synthesis and Characterization of 1,1-Di-Tert-Butyldiazene Citation McIntyre, Daniel Keith (1983) Synthesis and Characterization of 1,1-Di-Tert-Butyldiazene. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/y54d-s069. https://resolver.caltech.edu/CaltechTHESIS:12022013-142422017 Abstract The synthesis and direct observation of 1,1-di-tert-butyldiazene (16) at -127°C is described. The absorption spectrum of a red solution of 1,1-diazene 16 reveals a structured absorption band with λ max at 506 nm (Me 2 O, -125°C). The vibrational spacing in S 1 is about 1200 cm -1 . The excited state of 16 emits weakly with a single maximum at 715 nm observed in the fluorescence spectrum (Me 2 O:CD 2 Cl 2 , -196°C). The proton NMR spectrum of 16 occurs as a singlet at 1.41 ppm. Monitoring this NMR absorption at -94 0 ± 2°C shows that 1,1-diazene 16 decomposes with a first-order rate of 1.8 x 10 -3 sec -1 to form isobutane, isobutylene and hexarnethylethane. This rate is 10 8 and 10 34 times faster than the thermal decomposition of the corresponding cis and trans 1,2-di-tert-butyldiazene isomers. The free energy of activation for decomposition of 1,1-diazene 16 is found to be 12.5 ± 0.2 kcal/mol at -94°C which is much lower than the values of 19.1 and 19.4 kcal/lmole calculated at -94°C for N-(2,2,6,6-tetramethylpiperidyl)nitrene (3) and N-(2,2,5,5-tetramethylpyrrolidyl)nitrene (4), respectively. This difference between 16 and the cyclic-1,1-diazenes 3 and 4 can be attributed to a large steric interaction between the tert-butyl groups in 1,1-diazene 16. In order to investigate the nature of the singlet-triplet gap in 1,1-diazenes, 2,5-di-tert-butyl-N-pyrrolynitrene (22) was generated but was found to be too reactive towards dimerization to be persistent. In the presence of dimethylsulfoxide, however, N-pyrrolynitrene (22) can be trapped as N-(2,5-di-tert-butyl-N'-pyrrolyl)dimethylsulfoximine (38). N-(2,5-di-tert-butyl-N'-pyrrolyl)-d 6 -dimethylsulfoximine (38-d 6 ) exchanges with free dimethylsulfoxide at 50°C in solution, presumably by generation and retrapping of pyrrolynitrene 22. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Chemistry) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Goddard, William A., III Thesis Committee: Dervan, Peter B. (chair) Goddard, William A., III Grubbs, Robert H. Dougherty, Dennis A. Defense Date: 1 October 1982 Funders: Funding Agency Grant Number NSF UNSPECIFIED Record Number: CaltechTHESIS:12022013-142422017 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:12022013-142422017 DOI: 10.7907/y54d-s069 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 8035 Collection: CaltechTHESIS Deposited By: INVALID USER Deposited On: 02 Dec 2013 22:43 Last Modified: 06 Aug 2025 23:59 Thesis Files Preview PDF - Final Version See Usage Policy. 10MB Repository Staff Only: item control page

'lbesis by Daniel Keith Mcintyre In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1983 (Submitted October 1, 1982} ii To My Parents and to SUe iii Acknowled9ai!P11ts I would like to thank my research advisor, Peter Dervan, for his enthusiasm and support throughout this work. I would also like to thank the merbers of the Dervan group for contributing to a friendly, cohesive atmosphere which isn't always found in a research group. This especially applies to Dan Duan and Mark Mitchell whom I've had the pleasure of knowing from start to finish. Special thanks go to Bill Hinsberg for all his help in my first year here and for laying the ground work in the study of persistent 1,1-diazenes. I am indebted to Jan Owen and Debbie Chester for getting this thesis typed during a busy time. Thanks also go to the staff of the Caltech Chemistry Department for their frequent assistance. Bill Croasmun' s conscientious efforts in obtaining 500 MHz spectra at -120°C are greatly appreciated. Financial support of this research from the National Science Fcxmdation is also acknowledged. Final thanks go to sue for her love and supPort which made it all easier. iv Abstract '1be synthesis and direct observation of 1,1-di-,tell- butyldiazene (16) at -1270C is described. The absorption spectrum of a red solution of 1,1-diazene 16 reveals a structured absorption band with Amax at 506 nm (M~O, -1250C). about 1200 cm-1. The vibrational spacing in S1 is '!be excited state of 16 emits weakly with a single maximum at 715 nm observed in the fluorescence spectrum (Me20:~Cl2, -1960C). The proton ~ spectrum of 16 occurs as a singlet at 1.41 ppm. Monitoring this NMR absorption at -940 % 20C shows that 1,1-diazene 16 decomposes with a first-order rate of 1.8 x l0-3 sec-1 to form isobutane, isobutylene and hexamethylethane. This rate is 108 and 1034 times faster than the thermal decanposition of the corresponding .ill. and trans 1,2-di-,tell-butyldiazene isaners. The free energy of activation for decomposition of 1,1-diazene 16 is found to be 12.5 ~ 0 .2 kcal/IOOl at -940C which is ruch lower than the values of 19.1 and 19.4 kcal/lmole calculated at -940C for N-(2,2,6,6tetramethylpiperidyl)nitrene (3) and N-(2,2,5,5tetramethylpyrrolidyl)nitrene (4), respectively. This difference between 16 and the cyclic-1,1-diazenes 3 and 4 can be attributed to a large steric interaction between the ,tell-butyl groups in 1,1-diazene 16. In order to investigate the nature of the singlet-triplet gap in 1,1-diazenes, 2,5-di-,tell-butyl-N-pyrrolynitrene (22) was generated but was found to be too reactive towards dimerization to be persistent. In the presence of dimethylsulfoxide, however, N-pyrrolynitrene (22) can be trapped as N-(2,5-di-.tertbutyl-N'-pyrrolyl)dimethylsulfoximine (38). N-(2,5-di-,tell- v butyl-N'-pyrrolyl)-d6-dimethylsulfoximine {3~} exchanges with free dimethylsulfoxide at 500C in solution, presumably by generation and retrapping of pyrrolynitrene 22. vi Page A. Theoretical Studies...................................... 1 B. Persistent Cyclic 1,1-Diazenes........................... 3 c. Acyclic 1,1-Diazenes..................................... 8 D. 11 The Singlet-Triplet Gap and N-Pyrrolynitrene............. RPSJT:I'S All> DJS:liR<j'J"OO Chgpter l; Acyclic 1,1-Diazenes............................ 13 A. Synthesis of 1,1-Di-~-butyldiazene (16).............. 14 B. Electronic Absorption and Emission Spectra of 1,1-Di-~-butyldiazene C. (16).......................... Proton NMR Spectroscopy of 1,1-Di-~butyldiazene (16)...................................... D. 16 21 Decomposition Kinetics of 1,1-Di-~butyldiazene (16}...................................... 24 E. Analysis of ProductS from 1,1-Di-~buty1hydrazine ( 24) Oxidations. • • • • • • • • • • • • • • • • • • • • • • • • F. Synthesis and Irradiation of 1,1,4,4Tetra-~-buty1-2-tetrazene G. 29 (28)...................... 36 Approaches to the Synthesis of 1,1-Di(1-adamanty1)diazene (17) ••••••••••••••••••••••••••• 39 vii Olapter 2; Generation and Trgpping of 2,5-Di- tert-butyl-~rolYloitreoe A. <22>...................... 43 Synthesis of N-Amino-2 ,5-di-.tet.tbutylpyrrole (33) ••••••••••••••••••••••••••.••••••••••• 44 B. Generation and Trapping of 2,5-Di-~ butyl-N-pyrrolynitrene (22) •••••••••••••••••••••••••••• 45 c. Thermal Chemistry of N-(2,5,-di-.tet,t.-butyl-N'pyrrolyl)dimethylsulfoximine (38) •••••••••••••••••••••• 47 ~~<E ••••••••••••••••••••••••••••••••••••••••• 49 ~- ...•...........................•...........••.•.•. 72 1 1,1-Diazenes (aminonitrenes, ~nitrenes) 1,1 unlike their rore stable 1 ,2-diazene isaners (azo carrpounds) 2,2 are usually not isolated or detected by spectroscopic methods but rather are assmned intermediates oo the basis of a substantial body of chemical evidence.l The 1,1-diazene can be represented . R R 1a 'N-N: / .. R / .. / N=N 1b R .. 2 by two different resonance structures, la and lb, and the relative :inqx>rtance of each is de:pendent on the substi tuents R. 1,1-Dialkyldiazenes, best represented by structure la, are isoelectronic to ketones. However, when the internal nitrogen lone pair is delocalized onto the substituents the diazene is less able to form the dipolar resonance structure la and is rore correctly described by structure lb. Such electron deficient 1,1-diazenes should exhibit nitrene-like behavior. '.ftleoretical studies The nature of the bonding and the relative energies of the states for the parent 1 ,1-diazene (H2~N) have been the subject of several theoretical studies.3 Recent GVB-CI calculations by Davis and Goddara3e indicate that structure 1 is an accurate picture for the ground state. '!be electron configurations for the three lowest 2 valence states are shown in Figure 1 with the calculated relative energies. H~ HFQ 50.7 kcal Hhl< HFQ t H?-§• 13.8 kcal ~ •• H Figure 1. GVB-CI calculations3e for a2N-N. The singlet ground state (S0 ) is planar with a large degree of delocalization of the internal nitrogen lone pair onto the terminal nitrogen. The GVB rr orbitals are similar to those for a double bond and the calculated N-N bond length of 1. 25 .R is equal to the value observed for diimide. The calculated dipole moment of 4.0 D also indicates substantial double bond character. The S1 and Tl states have electron configuration similar to the lowest singlet and triplet states of nitrene (H-N). Interaction of the amino nitrogen lone pair with the nitrene rr electron is equivalent in both spin states so the 37 kcal separation corresponds to the singlet-triplet gap for nitrene. The S1 excited state can be reached by a n-+ rr * A visible absorption near 560 rnn is transition from So. predicted fran the energy spacing between these states. The most 3 favorable geometry for both the sl and TJ. states is pyramidal due to antibonding character in the " bond. Persistent Cyclic 1,1-Diazenes. Direct experimental studies of the 1,1-diazene functional group became available recently with the preparation, by Hinsberg, Schultz and Dervan, of the persistent cyclic 1,1-diazenes J4 and 4S. Both molecules 4 3 are made from oxidation of the corresponding lramino c::arpounds by .te.t,t-butyl hyp:>chlorite at -780C. These diazenes CMe their persistance to their synthesis at low temperature where the loss of nitrogen does not occur and to their steric bulk which retards dimerization to form 2-tetrazenes. As expected from the theoretical studies, 1,1-diazenes 3 and 4 are colored compounds giving purple and red solutions, respectively. Absorption and emission spectra are shown in Figures 2 and 3. 6 absorptions are 11> "* The transitions which show a blue shift in isopropanol and have an extinction coefficient of 20. The emission occurs from s1 with a fluorescence quantum yield of 7xlo-3.7 The lifetime of the excited state S1 is about 23 nanoseconds with internal conversion being the primary mode of decay.? 4 Infrared absorptions are observed at 1595 an-1 for the 6 merrbered ring diazene and at 1638 an-1 for the 5 membered ring.6 These absorptions were characterized as the N=N stretch by measuring the shift produced when the diazene is labeled with 15N at the terminal nitrogen. The similarity of these values to the 1576 an-1 Raman active !rN stretch observed for trans 1,2-diazenes further stresses that the 1,1-diazene contains a large amount of double bond character. electronic ?(.- "-/.=~: 9 +- .. N=N: infrored ( cm"1) (nm) Xmaa(iPrOH) 14N:!,4N 14N:!,SN 543 526 1595 1568 497 487 1638 1612 Xmaa(CH2CI2) Warming the 1,1-diazene solutions to ooc leads to the decomposition products shown in Figure 4.6 Bimolecular decomposition (k 2> is a low A-low Ea process (Ea=6. 4+0 •9 kcal/trole and Log A:3 • 8±0 •7 for dimerization of 3) which predominates at temperatures below -300C. At higher temperatures the unimolecular loss of nitrogen becomes daninant. The Arrhenius parameters determined for the unimolecular 1,1-diazene decanpositions in ether are Ea = 20.0 ~ 0.4 kcal/trol and log A= 13.7 ~ 0.3 for 1,1,-diazene 3 and Ea = 19.0 ~ 0.6 kcal/trol and log A= 12.4 + 0.4, for 1,1-diazene 4. The direct and triplet sensitized photolyses of 4 yield tetrazene 14 and the four hydrocarbons l(r-13 in 4:1 and 9:1 ratios, 5 respectively.B A spin correlation effect is observed for the hydrocarbon products resulting from direct and triplet sensitized decomposition of d,l-15.7 ,,~ + N=N 15 In addition, the 1,1-diazenes 3 and 4 have been studied by la, 13c and lSN NMR spectroscopy.6,9,10 6 /\ I \ I I \ \ I I I I \ :: \ •• c \ I I £ \ . § \ ..,e \ \ 420 Figure 2. 450 480 \ ' 780 .......... ......... Ill() 1140 Absorption ( - ) and fluorescence (- - -) spectra of 3 in CFCl3 at -780C and -1960C, respectively.? r, II \ 1 I ' l ] lu " I\ \ I \ \ \ I I \ I \ I \ I 'j ' I I \ ' \_,"' " "'-... 41() 440 470 ~ ~30 560 ~ 680 710 740 770 1100 . . . .1....111 '""'' Figure 3. Absorption ( - ) and fluorescence (- - -) spectra of 4 in CFC13 at -780C and -1960C, respectively. 7 7 9 ¢ ¢ 12 13 14 Figure 4. Unimolecular and Bimolecular decomposition products for 3 and 4.6 8 Acyclic 1,1-Di.azenes Direct observation of an acyclic 1,1-diazene has not been reported. The requirement for steric protection in persistent 1,1-diazenes suggests 1,1-di-~-butyldiazene 16, 1,1-di-(1-adamantyl)diazene 17, and 1,1-di(l-norbornyl)diazene 18 as synthetic targets. These three 1,1-diazenes are an interesting series because of their relationship to the bridgehead substituted 1,2-diazene isamers.ll,l2 N II :N .. ~~~ .. ~.~ N 16 17 18 x.x :N II N: .. Bridgehead substitution in traos-1,2-diazenes greatly increases stability towards fragmentation. Relative rates for extrusion of nitrogen from the syrrmetrical .tell-butyl, 1-adamantyl and 1-norbornyl traos-1,2-diazenes are listed in Figure s.ll This stabilization effect is also observed in the d.a-1,2-diazenes produced by irradiation of the trans isomers.l2 9 Irons- I ,2-diozenes 1. 0 N:1:2 -6 5 X 10 Figure 5. Relative decarq;x>sition rates for bridgehead substituted trans 1,2-diazenes.ll 10 '!be steric interaction of the .te.ct.-butyl groups becanes quite severe in lli.-1,2-di-.te.ct.-butyldiazene. The enthalpy of activation for decomposition of the ill. isomer is 18.6 kcal/molel3 compared to 43.1 kcal/molel4 for the trans isomer. Only part, probably about half, of the 24 kcal of strain energy in lli.-1,2-di-.te.ct.-butyldiazene comes from the ~-butyl steric interaction with the remainder due to electronic repulsion of the ill.- nitrogen lone pairs. The severity of the lone pair interaction in the crowded ~1,2-~-butyldiazene is indicated by the Amax at 447 mn which is 4000 cm-1 lower in energy than the Amax (380 mn) for ~1,2-di-isopropyldiazene.l2a The bridgehead lli.-1,2-diazenes also have large steric interactions between the tertiary groups. Because bridgehead substitution slows down fragmentation, however, isomerization to the trans isomer is competitive with loss of nitrogen for W.-1,2-di(1-adamantyl)diazene and is the sole decomposition process for lli.-1,2-di(l-norbornyl)diazene.l2 The bridgehead stabilization effect on 1,2-diazenes is apparently not entirely due to the instability of the bridgehead radicals which are for.med on decomposition. Several measures of radical stability agree that stability of adamantane-1-yl is nearly normal for a tertiary radical,lS yet trans-1,2-di(l-adamantyl)diazene decomposes about 3500 times slower than the ,tell-butyl analog .11 This peculiarity in the azo decompositions has been attributed to a requirement for planarity in the transition state which is even greater than demanded by the developing radical center.l6,11 It would be interesting to know if 1,1-di(l-adamantyl)diazene 17 is stabilized 11 in a similar way. Acyclic 1,1-diazenes containing tertiary substituents are expected to have destabilizing steric interactions as do the .c.ia-1,2-diazenes. Bridgehead substitution, however, should counter the destabilizing steric interaction by increasing C-N bond strengths. A diazene such as 1,1-di(l-norbonyl)diazene (18) might even be stable at room temperature if the steric protection is sufficient to prevent dimerization. Chapter 1 of this thesis describes the synthesis and characterization of 1,1-di-~-butyldiazene (16) along with studies directed towards the preparation of 1,1-di(l-adama.ntyl)diazene (17). 'Dle Singlet-Triplet Gap and N-Pyrrolylnitrene The lowest triplet state (TJ_) for 1,1-diazene (H2~> is predicted from theoretical studies to be about 14 kcal/mole higher in energy than the singlet ground state {SQ).3e The isoelectronic carbonyl functional group, on the other hand, has a small energy gap between TJ. and the first excited singlet state (Sl) which makes Tl easily accessible photochemically. Reducing the availability of the amino lone pair for delocalization ooto the univalent nitrogen in a 1,1-diazene should destabilize the ground state singlet and could conceivably raise So to be nearly degenerate with or even higher in energy than the lowest triplet state. The triplet state of a properly designed aminonitrene, then, might be accessible as the ground state. N-pyrrolylnitrene (19) has a destabilized singlet state because the amino lone pair is delocalized into a pyrrole ring which has a resonance energy of about 21 kcal/mole.l7 The high energy nature of this molecule relative to 1,1-dialkyl diazenes is evidenced by the 12 stability of the sulfonylhydrazine sodium salt 20 which is stable at 22SOC18 while the dialkyl sulfonyl hydrazine salts 21 decompose to diazene products at llOOC.lb .. R, /Na N-N, R/ S0 2Ar N-N: .. 19 21 20 The trpyrrolylnitrene chosen for this study is 2,5-di-~­ butyl-trpyrrolylnitrene (22). The ~-butyl substituents are designed to provide steric bulk to protect 22 from dimerization in solution and from attack at the reactive a positions of the pyrrole ring. Although sane electron deficient 1,1-diazenes have been trapped with olefins and sulfoxides, trapping of a pyrrolylnitrene has never been reported.l The generation of 22 and trapping of this species with dimethylsulfoxide is described in Chapter 2. 22 13 RESULTS AND DISCUSSIOO Q)apter 1 Acyclic 1,1-Diazenes 14 Synthesis of 1,1-Di~Jdiazene (16). Synthesis of 1,1-di-~-butylhydrazine {24) is the first step towards generation of 1,1-di-~-butyldiazene {16). 1,1-di-~butylnitrosamine The synthesis of {23) {Figure 6) is known in the literature although reduction of this compound by standard methods to the hydrazine was reported to be unsuccessful.l9 We find, however, that reduction of 1,1-di-~-butylnitrosamine {23) with sodium in refluxing ethanol under an inert atmosphere produces hydrazine 24 in 55% yield. ~, ~ N-NO ~ ~, Na .... EtOH ~ N-NH ?\2 24 23 Hydrazine 24 is extremely air sensitive and must be purified immediately before use by preparative VPC with collection under an argon atmosphere.20 We presume that steric interaction between the .te..t.t.-butyl groups causes this sensitivity to oxygen. Abstraction of a hydrogen atom would afford a stabilized hydrazyl radical 25, with ~N-NH __. ~.N-N. ?\2 .... .... 24 ?\ .... ~N-N , ~ ... + ?\ 'H 25 'H 15 0 z I >(ZY, ~~ 0'1 l""'i ~ :I: >(Y, ~ "ij Ul 0 ~ +J ..-4 c: l""'i ~ B ;t~ i z z ......I - l""'i -- ? N ' 0 N >(z l""'i ~ 0 Ul ...... Ul Q) :5 ~ • d N :I: >(z IQ ~ ~ 16 delocalization to form a 3 electron bond, increasing the angle between the ~-butyl groups. Moreover, relief of steric interaction should also occur when hydrazine 24 is oxidized to the 1,1-diazene 16. Addition of ~-butylhypochlorite to a dimethylether solution of 1,1-di-~-butylhydrazine {24) and triethylamine at -780C produces a white precipitate without the appearance of color indicative of a persistent 1,1-diazene. However, when the oxidation is performed at -1270C using a n-propanol/liquid nitrogen bath a Led solution which decolorizes rapidly at -900C is obtained in addition to the white precipitate. Darker solutions are formed when a precooled solution {-1270C) of ~-butylhypochlorite in chlorotrifluoranethane is added in portions to the rapidly stirred dimethylether solution of 1,1-di-~-butylhydrazine (24) and triethylamine. Filtration of this colored solution at -1270C to remove triethylamine hydrochloride gives a clear red solution of 1,1-diazene 16 for spectroscopic analysis. The bulk of the chlorotrifluoromethane is removed under vacuum to increase the concentration of the solution. ~~ ~ N-NH -,\2 24 :::<. 7\ . N=N: 16 Electrali.c Absorption and Fmission Spectra of 1,1-Di~ The 1 ,1-diazene functional group is isoelectronic to a carbonyl and exhibits a n _. " * electronic transition in the visible. The 17 visible absorption spectrum of 1,1-di-,tell-butyldiazene (16) in di.methylether was obtained using a low tanperature spectroscopic cell kept at -1250C with frequent additions of liquid nitrogen and is shown in Figure 7. The.\max at 506 nm for 1,1-di-~butyldiazene (16) falls between the A max of the cyclic diazenes as does the corresponding ketone series shown in Table 1. As expected for a rr>rr* transition where the ground state is rore polar than sv the use of a more polar solvent, di.methylether: n-propanol (1:1), shifts the spectrum 730 on-1 to higher energy. The fine structure observed in the absorption spectrum shows vibrational spacings of about 1200 on-1. This corresponds to the stretching frequency for the nitrogen-nitrogen bond in sl which has been weakened by antibonding character. The nitrogen-nitrogen double bond in the ground state of 1,1-di-~-butyldiazene (16) should have a stretching frequency of about 1600 on-1 based on stretching frequencies observed for the cyclic diazenes 3 and 4.22 The excited state of 1,1-di-,tell-butyldiazene (16) emits weakly at -1960C with a Amax at 715 nm as shown in Figure 8. The width of the emission band is large enough to contain up to three unresolved vibrational spacings of 1600 cm-1 for the ground state. The absorption and fluorescence Amax are separated by 209 nm indicating that the equilibrium geometries of So and s1 states of 1,1-diazene 16 differ substantially. 18 ,,,-,\ ,, l' I 1 c: -... 0 Q. I I I I I II I '\ ,, ' I '\ \ \ '\\ ~ \\ \ \ 0 en \, ', .0 <t \ \ \ \ ,, ' ' 600 500 Wavelength (nm) Figure 7. Absorption spectrum of 16 at -l250C in Me20 {-) and n-PrOH:Me20 (l:l) (-- -). :§ w .. u .~ u 0 '; .. ;g f c Figure 8. 500 Wovelen9lh (nm) ,. 700 \ \ \ \ \ \ • Emission spectrum of 16 at 770K in Me20-(l)2Cl2 glass (-·-·-). and n-Pr00-Me20 (- - -). '\ • '•' \• 800 \ ' '• • '' Absorption spectrum of 16 at -1250C in M~O (-) 600 . i I. .I. I I I I. . ! I I I . .. .I i I i i I .,.,·-· . . ., . ,-' '\ \ • '' ' • \\ .. ·w..e c .2 s• •c f !': ...... 1.0 N = ti= - .. N=N: + Q 7\ , :::{+ - rX· y·N: ~. 543 nm 506 nm 497 nm A 18416 19763 20120 E(cm- 1 ) 32830 33650 33725 E(cm- 1 ) max A i· and 16 and Their Corresponding Ketones. Comparison of Absorption Maxima for 1,1-Diazenes max (Me 20) TABLE I. 305 nm 297 nm 296 nm (hexane) Q=O 7\ C=O :::{, Q=O 1\.l 0 21 Proton NMR Spectroscopy of 1,1-Di--.tert--butyldiazene (16). Proton NMR spectra of 1,1-diazene 16 were ootained at -1200C using dimethylether: deuteriodichlorornethane (4:1) as solvent.23 Separate oxidations produced nearly identical solutions with a representative spectrum shown in Figure 9. The singlet at 1.41 ~ is assigned to 1,1-di-.te.tt-butyldiazene (16) and disa~rs along with the red color when the sample is warmed to -800C. The ranaining absorptions in the spectra are acetone as internal reference (s, 2.1 ppn), isobutylene (s, 1.68), .tm-butylchloride (s, 1.58), triethylamine (t, 0.97), isobutane (d, 0.87) and hexarnethylethane (s, 0.84). The peak at 1.16 ppn is canposed of unreacted hydrazine 24, ~-butanol and possibly 1,1,4,4-tetra-.te.tt-butyl-2-tetrazene. One of these, probably .tm-butanol, shows a temperature dependence for chemical shift which causes the 1.16 ~ absorption to split at -800C. ~.N=N: _J .. .{~ )= 16 As the 1,1-diazene signal at 1.41 ppm disappears, increases in the hydrocarbon absorptions can be observed by cut and weigh integration. Results for several sets of spectra indicate that the 22 Figure 9. Low tenperature proton ~ spectra a) of a red 1,1-di-~­ butyldiazene (16) solution at -1200C and b) after warming the solution at -aooc. 24 increases in the three hydrocarbon products account for the amount of diazene decanposed and that isobutylene and isobutane are formed in a ratio of approximately 1:1. The ~-butylchloride and large amount of isobutylene present in the -1200C spectra ~rently arise by a pathway other than 1, 1-diazene decanposi tion and will be discussed later . Deo-.:l()Sition Kinetics of 1,1-Di.-t;ert-but.yJdiazene (16). The decay kinetics of 1,1-diazene 16 were studied by monitoring the disa~arance of its NMR absorption at four temperatures between -940C and -1070C (Figure 10) • The samples used for kinetics were the same as those described for NMR spectroscopy. Since only one of the four temperatures falls on the methanol temperature calibration curve24, temperature calibration was a problem. calibration of the lOOC range above -940C was assumed to hold for the region -94°C to -1050C and error in temperature measurement was estimated to be within ±.2oc. 25 Disa~arance of 1,1-diazene 16 is first order with no evidence of curvature through as many as four halflives in the plots of ln A versus time shown in Figure 11. The rates for each temperature are listed in Table 2 and were determined by least squares analysis. Free energy of activation ( ~G*) was calculated for each temperature,26 but the errors involved do not permit accur ate determination of the Arrhenius parameters. An Ea of 9. 7 +4.3 kcal/mol27 and a log A of 9.1 ~ 5.5 are obtained from least squares analysis of the ln k versus l/T plot for the four rates.28 Since the activation parameters for the 1,2-diazene isomers are 25 a) Figure 10. Proton NMR absorptions at -940C for a) 1,1-di-~­ buty1diazene (16) and b) ~-buty1chloride measured every 114 seconds. 26 w -102 z w ° N <( 0 '-- c -2.0 _______________.2_______________3~----- -3.0~------------~1 TIME (1000 SECONDS) Figure 11. Ln A versus time plots for decanposition of 16 in dimethylether. 27 known, relative rates for thecnal decanposition can be calculated for carp:lrison with 1,1-diazene 16 at -940C (Table 3). The gap between the relative rates of the~ and trans isaners is very large because the ~ roolecul.e is destabilized by a steric interaction between the .t,ert-butyl groups and severe electron repulsion in the IL. lone pair roolecul.ar orbital. The 1,1-isaner 16, which should be inherently less stable in the strainfree case, is further destabilized by the .tell-butyl steric interaction.29 The differences in the free energies of activation (~G*> for the 1 ,1-diazene decanpositions can be interpreted as relative strain energies if the 1, 1-diazenes are decarp:>sing by similar mechanisms with similar entropies of activation ~s*) and if most of the strain is relieved in the transition state. On this basis, 1,1-di-.tett,-butyldiazene (16) has about 7 kcal/mole more strain energy than do the cyclic 1,1-diazenes 3 and 4 which have nearly equal amounts of strain. The strain energy for 1,1-di-~-butyldiazene (16), then, is 7 kcal/role plus whatever strain energy exists in the cyclic diazenes. Since the strain energy in the cyclic diazene3 is expected to be small, the strain in 1,1-di-~-butyldiazene (16) appears to be consistent with the strain energy of 12 kcal/mole calculated for the _tett-butyl interaction in l,l-di.tett,-butylethylene.25 28 ~1e 2. First Order Rate Constants for Decomposition of 16. Temperature (OC) No. of spectra -94 20 57 1.77xlo-3 -97 26 120 o.95xlo-3 -102 5 240 o.79xlo-3 -107 5 600 O.l7xl0-3 Table 3. time interval (sec) k1 (sec-1) Relative Rates for Decomposition of Di-~-butyldiazene Isomers at -940C. Diazene Solvent traos-1,2-di-~-butyl hexadecane 40.2a 2xlo-34 ~-1,2-di-~-butyl pentane 17.7b Sxlo-7 dilrethylether 12.5 1 1,1-di-~-buty1 (16) a) Ref. 14: b) Ref. 13. krel 29 'nlbl.e 4. Relative Rates and ~G*'s for 1,1-Diazene Decarpositions at -940C Diazene ~G* (kcal/tool) 3 19.la krel 4 12.5 16 1 9) Ref. 6. Analysis of Products frma 1,1-Di.~.lhydrazine (24) Oridatims. Products obtained from the reaction of 1,1-di-~-butylhydrazine (24) with two different oxidants, ~-butylhypochlorite and nickel peroxide, were analyzed by analytical VPC (Table 5) • were carried out at -7SOC. Both reactions 1,1-Diazene 16 is not stable at this temperature. Table 5. Product Ratios From Oxidation of 1,1-di-~-butylhydrazine (24) .a 11)-H >= X ~~)-c• ~ ~-Butylhypochlorite 1.0 2.6 0.1 Nickel Peroxide 1.0 1.6 0.2 Oxidant ~lar ratios deter.mined by VPC. ... )L N·~ 0.5 0.03 30 'lhe three hydrocarbon products, isobutane, isobutylene and hexamethylethane, are produced in molar ratios of 1.0:2.6:0.1 fran the hypOchlorite oxidation with a total yield of 38% based on C4 groups. A 5% yield of ~-butylchloride and a very small oomponent (0.3%), which may be 1,2-di-~-butyldiazene by VPC, were also observed from this oxidation. Oxidation with nickel peroxide, however, quantitatively produces only the eydrocarbon products. The molar ratios found for isobutane, isobutylene and hexamethylethane are 1.0:1.6:0.2. The 1,2-diazene isaner is not observed fran this oxidation. Monitoring the at:{)earance of the three hydrocarbon products by NMR while diazene 16 decanposes shows that the increases in the hydrocarbon absorptions account for the amount of diazene lost. Dimerization of 1,1- diazene 16 to for.m tetra-~-butyl-2-tetrazene (28) is therefore not a significant pathway for diazene decomposition. The hydrocarbon products are formed in an approximate molar ratio of 1.0:1.0:0.05 (isobutane:isobutylene:hexamethylethane) whi~ is consistent with molar ratios obtained fran VPC analysis of the bulk oxidation products i f excess isooutylene is explained as arising from other oxidative processes which will be discussed below. A reasonable mechanism for decanposition of 1,1-di-~-butyldiazene (16) involves loss of nitrogen by either one or two bond cleavage to give ~-butyl radicals which disprqx:>rtionate or canbine to give isobutane, isobutylene and hexamethylethane. Thermal and photochanical decanposition of the 31 1,2-di-~-butyldiazene isaners and di-~-butylketone give isobutane,isobutylene, and hexamethylethane in the amotmts shown in Table 6 consistent with the behavior of~ butyl radicals. The product ratios observed for 1, 1-diazene 16 by Mom are similar to the ratios reported for decomposition of ~-1,2-di-~-butyldiazene at -230C. .:y• -~' ~. • 7\ .. N=N: ~ -=-..:-~ 16 •• • _JN=N .. .{~ Figure 12a. Radical mechanism for decanposition of 1,1-diazene 16. 32 ~1e Molar Product Ratios for Sane Related Decanpositions 6. Canpound •1-H >= ~~H~~ Decanposition ~-1,2-Di-~butyldiazenea Me<E, -230C 1.0 1.0 0.1 traos-1,2-Di-~-butyldiazeneb hv , 300C 1.0 0.92 0.12 traos-1,2-Di-~-butyldiazenec 9 torr, 2600C 1.0 1.0 0.36 Di-~-butylketoneb hv 1 300C 1.0 0.87 0.15 a) Ref. 30, b) Ref. 31, c) Ref. 32. A second possible mechanism for decomposition of 1,1-diazene 16 is concerted elimination of isobutylene. The .t.e.tt.-butyldiimide which would be formed in this mechanism can decanpose by either a bi.Ioolecular or a radical , H -:;._/ ~.­ N=N: _J .. .{~ 2 ..N=N/ H + 7\ .. 16 Figure l.2b. (16). Concerted mechanism for decanposi tion of 1,1--diazene 33 chain pathway to give isobutane.34 The major products expected from a concerted decomposition of l,l-1:3iazene 16 are, then, isobutylene and isobutane. Evidence for such a mechanism has been found in the stereospecific elimination of 2-butene from 1-~tyl-1-phenyldiazene :i~ (26). . ).-( Q "-N I H " 26 Although the concerted mechanism cannot be ruled out for 16, the formation of hexamethylethane appears to indicate that loss of nitrogen to give .tetl,-butyl radicals is a decomposition pathway for 1, 1-di-.tetl,-butyldiazene (16) • As mentioned before, .tetl,-butylchloride and a large amount of the isobutylene formed in the hypochlorite oxidation apparently arise from some oxidative pathway other than decomposition of the 1,1-diazene. 'lbese two products suggest the intermediacy of .tetl,-butyl cation and can be explained by the mechanistic scheme shown in Figure 13. Formation of the diazenium ian from 1,1-hydrazines having a good leaving group in the 2 position is precedented by the exchange of sulfonate groups in 1,1-dialkyl-2-sulfonylhydrazines. .tetl,-butyl cation from Cleavage of 24 N-NH ,, ~ ... \ 2 t-luOCI~ '' 'H Cl _. 'H ~IJN 27 16 tt=ii= --' f\ .:::{ , '~ ~ Cl ,,N=N -__. ~. N =N'H :::{ Cl ... ~ I· I..OCI.. 7\ of 1,1-di-~-butylhydrazine (24). __. "" ~ Cl )= )= A·ri --+ '}. Cl N=N, / .:::{ ~ ~~ ~ A•H Figure 13. Mechanistic scheme for ~-butylhypochlorite oxidation ~... N-N' "~, ~ ~ / - . )-c• ~ w 35 1,1-di-~-butyldiazenium ion (27) relieves the steric interaction between the ,tell-butyl groops. The ~-butyl cation can either carbine with chloride ion to give ~-butylchloride or lose a proton to triethylamine to form isobutylene. ~&ltyldiimide is very susceptible to radical induced chain decarp)si tion and should also react with .m.t-butylhypochlorite. The trans-1,2-di-~butyldiazene which awea,rs to be a product of the oxidation could be formed by capture of mt.-butyl cation by ~butyldiimide followed by deprotonatioo. Nickel peroxide is a cleaner oxidant which unfortunately reacts too slowly at -1270C to generate concentrated solutions of 1,1-diazene. The high ratio of isobutylene to isobutane (1.6:1) may arise from reaction of ~-butyl radical with the excess of nickel peroxide used. The ratio of products observed is what would be expected from 77% cage disproportionation-cambination with complete scavenging of out of cage radicals by nickel peroxide to form isobutylene. Tanner, et al., find 50-90% cage reaction, depending on solvent viscosity, for thermal decanposition of ~-1,2-di-mt,-butyldiazene generated at 300C from photolysis of the trans isaner.31 36 (28). Biroolecular reaction to form 2-tetrazenes has been observed for the cyclic 1,1-diazenes but is too slow to caopete with the increased rate of unimolecular decomposition in 1,1-di-~-butyldiazene (16). 1,1,4,4-tetra-~-butyl-2-tetrazene (28} can be made, however, by stirring neat 1,1-di-~-butylhydrazine (24} at OOC under an atmosphere of oxygen. Hydrazyl radical 25 and 1,1,4,4-tetra-~-butyltetrazane (29}, shown in Figure 14, are the probable intermediates in this reaction. White needles are produced which give correct elemental analysis and are very similar to those obtained for 1,1'-azo-2,2,5,5-tetrarnethylpyrrolidine (14).6 28 14 9 37 crt • CIO N )lz)< z' z 'rz)< I % ~ / / I ~ N 10 w ~ ~ ~ 0 § "" +J ~ 0 ~ w 0 ~ t m "" z 'rz)< / ~ I ot z 'rz)< N X I ~ ~ .... .-4 4J ~ 38 Unlike tetrazene 14, however, 1,1,4,4-tetra-~-butyl2-tetrazene ( 28) is labile in solution at room tanperature and even solid samples sometimes decanpose on standing at roan temperature. A deuteriobenzene solution of tetrazene 28 sealed in a l'l-tR tube decanposes over a few days at room temperature to form di-~-butylamine and the .te.r.t.-butyl radical products, isobutane, isobutylene and hexamethylethane. The mechanism for formation of these products is unknown but may be related to attack of dialkylamino radical on the 2-tetrazene which has been reported for 1,1'-azo-2,2,6,6-tetramethylpiperidine (9).35 Radical induced chain decomposition could explain the variable stability of solid samples of tetrazene 28. The UV spectrum of tetrazene 28 has a single absorption at 230 nm ( ( 1500) in diethylether which is in sharp contrast to the double maxima at 254 nm and 300 nm observed for tetrazene 14. Dimethylether-dichloromethane and 2-methyltetrahydrofuran solutions of tetrazene 28, saturated at OOC, can be quickly frozen at -1960C to give clear glasses without precipitation of tetrazene. When these glasses are irradiated with UV light a red color is produced within 15 minutes but continued irradiation does not produce an observable increase in intensity. When the red glass is thawed at -1270C (n-propanol/liquid nitrogen) bubbling occurs and tetrazene 28 precipitates from solution while the red color persists. Further warming above -sooc destroys the red color with gas evolution and the resulting colorless solution can be refrozen and irradiated to reform the red coloration. Attempts to obtain clear red solutions free of precipitate for optical and NMR spectroscopy failed and the red 39 conpxmd was never characterized. From these observations, the red substance appears to be a photolabile intermediate which is generated at a steady state concentration deteonined by the concentration of tetrazene 28. The appearance and stability of the red color are similar to what is observed for 1,1-di-.tet,t-butyldiazene (16) although cleavage of the tetrazene into its diazene monaners is considered unlikely. possible intermediate is the ill. isaner of tetrazene 28. Another Ingold, et. al.,36 have reported the observation of a photolabile intennediate in the photolysis of 1,1,4,4-tetra-isopropyl-2-tetrazene which they suggest is ill,-1,1,4,4-tetra-isopropyl-2-tetrazene. This intermediate forms diisopropylamino radicals either !=hotochenically or thermally with an Ea of about 24 kcal/roole. Based on these observations, it is conceivable that ill,-tetra-.tet,t-butyl-2-tetrazene, like 1,1-diazene 16, could be a red canpound which is thermally labile at -ao<>c • .AWroaches to the Synthesis of 1,1-Di(1-adamanty1)diazene (17). The possibility of stabilizing acyclic 1,1-diazenes with bridgehead substitution makes 1,1-di(l-adamantyl)diazene (17) an attractive synthetic target. 1-Adamantyl is the only bridgehead group for which the synthesis of the disubsti tuted secondary amine has been reported.37 Di(1-adamantyl)amine (30) is needed to make the diazene precursor, 1,1-di(l-adamantyl)hydrazine (31). 40 '!he overall yield of the secondary amine in this synthesis (Figure 15), however, is only 5% and a more efficient route was desired. Na Ll )[}., Figure 15. 30 Literature route to di (1-adamantyl) amine.37 We found that heating two equivalents of 1-adamantylamine with 1-adamantylbranide at 2200C produced di-1-adamantylamine (30) in 80% isolated yield. Nitrosation of this amine with nitrosylchloride in pyridine gives 1,1-di(l-adamantyl)nitrosamine (32). This nitrosamine, like 1,1-di-.t,ett-butylnitrosamine (23), is acid sensitive and partial decomposition to give 1-adamantanol and di-1-adamantylether is observed on silica gel. >2oc:f NOCI_. 30 32 41 Reduction of nitrosamine 32 to give 1,1-di(l-adamantyl)hydrazine (31) is difficult because of steric hindrance and sensitivity to rearrangement. Several basic reducing conditions were attempted under an inert atmosphere and amine products were collected by precipitation as hydrochlorides. In every case the only amine product found was the overreduced di(l-adamantyl)amine (30). Lithium in liquid arrmonia gave a small amount of secondary amine with the rest of nitrosamine 32 left unreacted. Allmlinum amalgam in refluxing diethylether ,sodil.Dll in refluxing ethanol and lithil.Dll aluminum hydride in di-n-butylether at lOOOC all produced 1-adamantanol and di-1-adamantylether along with the overreduced secondary amine. The lithium allmlinum hydride reaction also yielded sane adamantane and 1,1'-biadamantane. These radical products may result from decomposition of a reduction intermediate or from oxidation of the desired hydrazine. '!be amine fraction from this reaction however shows no sign of either ·hydrazine 31 or the hydrocarbons which would be formed from its decanposition. From the results of this work with 1-adamantyl substitution it is concluded that the best approach for synthesizing a bridgehead substituted 1,1-dialkyldiazene is to use a smaller group such as l-bicyclo[2.2.2] or l-bicyclo[2.2.1] (1-norbornyl). Both of these tertiary bridgehead groups show a large stabilization effect in the 1,2-diazenes.ll The smaller size will lessen strain due to steric interactions in the 1,1-diazene and should make the nitrosamine easier 42 to reduce. '!he difficulty in these two systems, however, will be synthesizing the secondary amines. There is same reason to believe that the bicyclo[2.2.2] group can be used in a similar condensation reaction as was used to make di-l-adamantylamine.38 Di-1-norbornylamine may be approached by reaction of 1-norbornyl lithium with 1-nitronorbornane followed by hydrolysis to give di-1-norbornylnitroxide which could be easily reduced to the amine.39 43 RESULTS AND DISCUSS!~ Qmpter 2 Generation and Trapping of 2,5-Di-~-butyl-:trpyrrolylnitrene {22). 44 Synthesis of 1H'IIlil»-2r5-di-1;.e.t:t-butylpyrro1e (33) • N-.Amino-2,5-di-~-butylpyrrole (33) was desired as a precursor to 2,5-di-~-butyl-N-pyrrolylnitrene (22). The appropriately substituted pyrrole ring can be made by the 36 33 22 condensation reaction of 2,2,7,7-tetramethylocta-3,6-dione(34) with ammonia.40 The dione 34 is available from oxidative coupling of th~ enolate of pinnacolone using cupric chloride.41 The N-aminopyrrole 33 then, can conceivably be synthesized either fran direct condensation 35 34 of 34 with a hydrazine or by amination of pyrrole 35. Condensation of dione 34 with hydrazine gives 3,6-di-~­ butylpyridazine (36) as the isolated product after air oxidation. This is a convenient route for preparation of the ring expanded isoner of aminonitrene 22. Use of the blocked hydrazine, ~-butylcarbazate, gave coodensation to form the 5 ment>ered ring after a reflux period of 30 hours in glacial acetic acid. Hydrolysis of the carbazate occurred under the reaction conditions resulting in a 45 mixture of pyrrole and pyridazine products. N-amino-2,5-di-,tm- butylpyrrole (33) was isolated by chromatography and recrystallization in 40% yield. NH 2 NHC00t-Bu ~ , CH 3COOH 33 34 Generation and Trapping of 2,5-Di-.t.e.tt-but.yl-N-pyrrolylnitrene (22). N-Aminopyrrole 33 is quite resistant to oxidation at the arninonitrogen due to the reduced availability of the nitrogen lone For example, 33 is indefinitely stable in pair in the pyrrole ring. air, in sharp contrast to 1,1-di-.t&.t,t-butylhydrazine (24) which decanposes in minutes on exposure to air. Oxidants such as mercuric oxide, manganese dioxide and photochemically generated .t&,tt-butoxy radicals all gave no reaction with 33 at room temperature while halogenated oxidants such as iodine and .t&,tt-butylhypochlorite attacked the pyrrole ring. ~ad tetraacetate was the only reagent found to be suitable for oxidizing the amino group in 33. Reaction of N-arnino-2,5-di-.t&.t,t-butylpyrrole (33) with lead tetraacetate at OOC affords 1,1'-azo-2,5-di-~-butylpyrrolyle (37) as the only najor product. This reaction is very slow at -7BOC LTA 33 ... 37 46 producing a low yield of 37 after several hours with most of the aminopyrrole renaining unreacted. It appears fran this oxidation that pyrrolylnitrene 22 is being formed but is too reactive to be stable towards dinerization, even at -7a<>c. When the oxidation of 33 is performed in the presence of dinethylsulfoxide, pyrrolylnitrene 22 is efficiently trapped as N-(2,5-di-~-butyl-N'-pyrrolyl)dimethylsulfoximine (38). Dimethylsulfoxide is a nucleophillic trap which adds to electron 38 deficient aminonitrenes. The thermal behavior of this adduct will be discussed in the following section. Attempts to trap aminonitrene 22 with ethylene, tetramethylethylene and diethylfumarate were unsuccessful with dimerization of 22 to foon tetrazene 37 observed instead of olefin addition. A photochemical precursor to 1,1-diazenes would be very useful for making especially reactive roolecules in inert environments such as low temperature glasses. Preparation of stable N-azides has been attempted42 but the synthesis of dinethylaminoazide is the only successful example reported.43 N-azido-2,5-di-~- Photoextrusion of nitrogen from butylpyrrole (39) would be a convenient way to 47 generate pyrrolylnitrene 22 in rigid glasses or in a solution free of lead or acid contaminants. '!be synthesis of 39 was attempted using the diazo transfer reaction shown in Figure 16. A stable :t-razide was never isolated but observation of tetrazene 37 as a product along with bl.ltbling during the reaction were taken as evidence of an unstable 1-razide 39. 39 Figure 16. Attempted diazo transfer reaction from tosylazide to the lithium salt of aminopyrrole 22. 'lbermal Olemist.cy of !r- (2,5-Di--.text-butyl-N'pyrrolyl) 1-rPyrrolylsulfoximine 38 forms white crystals which are quite stable until heated near· their melting point at lllOC. In solution, however, :t-rpyrrolylsulfoximine 38 is labile at roam temperature. Heating 38 in tetrahydrofuran at SOOC produced 15% decomposition to tetrazene 37 and free dimethylsulfoxide in less than four hours. Further decomposition at this temperature is apparently inhibited by the free dimethylsulfoxide formed. To study this effect, :t-r{2,5-di-.tet.t,-butyl-N'-pyrrolyl)~-dimethyl-sulfoximine {38-t%) was prepared by lead tetraacetate oxidation of 1-raminopyrrole 33 in the presence of ~-dimethylsulfoxide. When deuterated sulfoximine 38-t% is heated with three equivalents of dimethylsulfoxide at SOOC in carbon 48 tetrachloride rapid exchange of the sulfoxide groups is observed. Analysis of the solution by NMR shows that the protonated DMSO lit dimethylsulfoxide is 89% equilibrated between adduct and free dimethylsulfoxide after 3 l/2 hours and equilibration is complete after 12 hours at SOOC. This exchange process at,:parently occurs by thermal generation of ~pyrrolylnitrene 22 followed by rapid retrapping with dimethylsulfoxide. The fact that dimerization occurs to form tetrazene 37 in the absence of dimethylsulfoxide supports this mechanism and rules out direct bimolecular substitution of sulfoxide groups. Synthesis of ~pyrrolylnitrene 22 from ~pyrrolyl sulfoximine 38 was attempted using flash thenoolysis with condensation of products at 77~. Heating 38 to sublime it into a quartz tube at 4500C produced decanposi tion in the sublimation tube with condensation of 2,5-di- .tett_-butylpyrrole 35 and dimethylsulfoxide. Pyrrole 35 is probably for.med from decomposition of tetrazene 37 which was collected along with pyrrole 35 and unreacted sulfoximine 38 when a solution of 38 was slowly dropped through a vertical quartz tube at 3600C under vacuum. No evidence for pyrrolylnitrene 22 at 77~ was observed from either of these pyrolyses. 49 Ezperimental. Section Melting points were determined using a Thomas-Hoover Melting Point ~ratus and are uncorrected. Infrared spectra were recorded on a Perkin-Elmer 257 or a Beckman IR 4210 spectrometer. Proton nuclear magnetic resonance {NMR) spectra were ootained on a Varian Associates EM-390 or a Jeol FX-90 spectrometer. Low temperature proton spectra were obtained using a Bruk.er WMSOO 500 MHz spectrometer. Cllemical shifts are reported as parts per million downfield fran tetramethylsilane { 5 units) and multiplicity is given as s=singlet, d=doublet, t=triplet, m=multiplet and b=broad. Electronic spectra were recorded using a Cary 219 or a Beckman Model 25 spectrophotometer. A uv-visible cell designed44 for use in conjunction with an Air Products WMX-lA Vacuum Shroud and LC-1-110 Cryo-tip Refrigerator, 45shown in Figure 17, was used to obtain electronic spectra at low temperatures. The windows for the vacuum shroud and the cell are Suprasil I, ground to size from blanks {Amersil, Inc.46). Samples were introduced into the cell through precooled 18 gauge teflon tubing {Alpha Wire Corp. 47) by applying suction with a syringe. co~r, '!be cell body is constructed from OFHC with stainless steel tubing soldered to it. The body is nickel plated and has a thermocouple well with set screw. Tbe seals between the windows and body were made with Viton ~Rings. The cell path length is 10 nm and the volmne is approximately 4 ml. '!be cell was constructed in the Chemistry Division/caltech Instrument Shop. Filtrations at low temperatures were performed using a medium frit filter disc built into a vacuum jacketed cooling bath. 'Ibis apparatus was constructed in the Caltech Glass Working Shop and is 50 Figure 17. Low temperature spectroscopic ce11 44 and shroud. 51 shown in Figure 18. A dry ice/acetone slush bath was used to cool to -7SOC and a liquid nitrogen/n-propanol slurry was used at -1Z70C. 1,2-Dimethoxyethane (glyme) was distilled fran lithium aluminum hydride. Ethyl ether and tetrahydrofuran were distilled from sodium benzophenone ketyl. Pyridine was dried over 4A JOOlecular sieves. Chlorotrifluoromethane and dimethylether were used directly as they were condensed fran gas cylinders. Deuteriodichloromethane was passed through a short colUim of basic alumina before use. was distilled from calcium hydride. Dichloranethane Dimethylsulfoxide was distilled from lithium aluminum hydride. Lead tetraacetate was recrystallized from acetic acid and p.mped dry. ~-Butylhypochlorite was washed with 10% sodium carbonate solution and water, dried (CaC12), and distilled under reduced pressure. KOH pellets. Triethylamine was distilled from 1,1-Di-~-butylhydrazine (24) was always p.Irified by preparative VPC (Pennwalt 223, 1300) and collected under argon atmosphere immediately prior to use. A Hewlett-Packard 5700 A gas chromatograph equipped with a Hewlett-Packard 1870 A inlet splitter and a flame ionization detector was used for capillary vapor phase chromatography (VPC) • Hydrogen was employed as the carrier gas and the makeup gas was nitrogen. Packed column analytical VPC was done using a Hewlett-Packard 5720 A gas chromatograph equipped with a flame ionization detector and nitrogen carrier gas. This instrument was used with 0.125 in. packed stainless steel columns. Quantitative VPC analysis with both instruments was accomplished using a Hewlett-Pac!<ard 3390 A electronic integrator. Detector response for hydrocarbons was assumed to be 1.0 relative to pentane. A Varian 920 instrument, equipped with a thermal 52 'J 14/20\ ,...;...-- 7.0 ••--+- 6 .0 -+-...1 MEDIUM FRITTED 'J 14/20·_ _ / DIMENSIONS IN CM Figure 18 . Vacuum jacketed low temperature filtration funnel . 53 conductivity detector, helium carrier gas and 0.375 inch packed aluminum colurtD'ls, was used for preparative VPC. '!he analytical and preparative colUJmS used are described in Table 7. Mass spectra were recorded on a DuPont 24-492B Mass Spectrometer. Elemental analyses were performed at Spang Microanalytical Iaboratory, Eagle Harbor, Michigan, and at Galbraith Iaboratories, Knoxville, Tennessee. Reactions were carried out under a positive pressure of dry argon. All low temperature solution transfers were performed using teflon tubing and a rapid stream of argon. 2-Methyl-2-nitroprqmne. The procedure of Kornblum was followed.48 Tb a stirred solution of 650 gm (4.11 nK>le) potassium permanganate in 3 liters of distilled water was added 100 gm (1.37 nK>le) of .tell-butyl amine. This mixture was stirred for 8 hours at roam temperature followed by heating to 550C for an additional 8 hours. '!he product was removed from the reaction mixture by st~ distillation at 900C. The organic phase was washed with 10% HCl solution and saturated sodium chloride solution, dried (Mg004), and distilled at 1270C to give 115 gm (80%) of a colorless liquid which solidified at roam temperature, 2-methyl-2-nitropropane: cm-1; NMR (CDCl3) IR (film) 3000, 2950, 1545, 1375, 1350, 860 o 1.6 (s). 54 Table 7 VPC Columns Column Designation Description Pennwalt-223 3'xl/4"glass, 28% Pennwalt 223 on 80/100 Chrom R. Pennwalt-223 4'xl/8" stainless steel, 28% Pennwalt 223 on 80/100 Chrorn R. Pennwalt-223 5'xl/4" glass, 28% Pennwalt 223 on 80/100 Chrom R. DBT 20'xl/8" stainless steel, 10% Dibutylytetrachlorophthalate on 100/120 Chrom P ~DMCS. Carbowax 400 lO'xl/8" stainless steel, 10% Carbowax 400 on 100/120 Chrom P ~DMCS SE-54 15 meter 0. 31 rrm ID fused silica SE-54 capillary. SF-96 l'xl/8" stainless steel, 10% SF-96 on 100/120 Chrom W ~DMCS. SF-96 lO'xl/8" stainless steel, 10% SF-96 on 100/120 Chrom P ~DMCS. 55 Di-J;.ert-tlut.yl.amine. Di-.tet,t-butylamine was prepared from 2-methyl-2-nitropropane by a modification of the procedure of Back and Barton.l9 Tb a 1 liter flask fitted with an overhead stirrer and containing 400 ml of freshly distilled glyme (LAH) was added 93 gm (0.90 roole) of 2-methyl-2-nitropropane. Addition of 21 gm (0.90 roole) of sodium cut into small pieces caused the surface of the metal to turn bright gold and the pale lavender color of the 2-methyl-2-nitropropane radical anion formed in solution. The suspension was stirred, as a white precipitate formed, for 4 days until the lavender color was no longer present. The white slurry was then added to a 5 liter flask containing 600 gm of crushed sodium sulfide nonahydrate and 12 gm (2.5 roole) of sulfur stirred from above in 900 ml of dirrethylformamide. 'Ille dark green mixture was illuminated from below with a 100 watt light bulb for 7 5 minutes. The reaction mixture was divided into two halves and each half was poured into 1 liter of a saturated potassium carbonate solution and extracted (4x250 ml.) with pentane. The canbined pentane extracts fran both halves were washed (2x200 ml.) with water and dried (Na2S04) • Gaseous HCl was bubbled into the pentane solution to precipitate the amine products. Tbe white precipitate was collected by filtration and washed with ether to give 49.2 gm of amine hydrochlorides. Shaking this solid with 20% NaOH solution liberated the amines and the organic layer was separated and distilled ( 60°-670C at 130 nm) to give 26 gm (45%) of di-~ butylamine (98% pure by analytical VPC, 15' Pennwalt-223,1300): 1365, 1230 cm-1~ NMR (CDCl3) o 1.17 (s). IR (film) 2980, 1475, 1385, 56 Di-J;ert-b.rt.ylnitrosamine (23). A modification of the procedure of Back and Barton was used.l9 A solution of 25 gm (0.19 mole) of di-~-butylarnine in 100 ml dry pyridine was cooled with an ice-methanol bath. Nitrosyl chloride was passed into this solution until a red color persisted. The cold mixture was added to 400 ml water and extracted (lx400 ml, 2x200 ml) with ether. The sensitivity of nitrosamine 23 to acid requires removal of pyridine by repeated washing of a pentane solution with water. The ether was removed under reduced pressure and the resulting orange oil was taken up with 400 ml pentane and 400ml water. The aqueous layer was separated and extracted (2x200 ml) with pentane. The combined pentane fractions were washed (4x300 ml) with water, dried (Na2S04) and concentrated under reduced pressure to give 26 gm (85%) of a yellow oil which solidified at room temperature, di-~-butylnitrosarnine (23): IR (film) 2970, 1435, 1390, 1370, 1265, 1190, 1120, 1015 am-1; NMR (CDC13 )o 1.67 (s, 9H), 1.52 (s, 9H); UV (Et20) 385 nm ( c 44), 236 (5300). 1 ,1-Di-.tert-b.rt.ylbydrazine (24) . Nitrosamine 23 was warmed until melted and 0.5 gm (3.2 mole) of the yellow oil was dissolved in 4 ml absolute ethanol. This solution was added to a 25 ml flask fitted with a reflux condenser under argon. As the flask was lowered into a preheated oil bath small pieces of sodium were added under argon flow. Additional pieces of sodium were added to maintain a mol ten globule of sodium in the refluxing ethanol for 20 minutes. During this time the yellow color had corrpletely faded and a white solid had formed . The mixture was then cooled in an 57 ice bath before 20 rnl of deoxygenated water was added to quench the excess sodium and dissolve the sodium ethoxide. The resulting cloudy solution was extracted (2x5 rnl) with ether in a separatory funnel which had been flushed with argon. The catiJined ether extracts were then concentrated to about 2 rnl under a stream of argon and dried (Na2004). Preparative VPC (3' Pennwalt-223, 1300) with collection under an argon atmosphere yielded 250 rng (55%) of 1,1-di-~-butylhydrazine (24) ( 99% pure by analytical VPC, SE-54): IR (film) 3380, 3240, 2980, 1485, 1390, 1370, 1205, 860 cm-1; NMR (CilCl3) 5 2.80 (b, 2H), 1.24 (s, 18H). 2-Benzami~1 ,1-Di-.tert-but:yl.hydrazine. To a suspension of 240 rng (1.67 rrmol) of freshly prepared 1,1-di-~-butylhydrazine (24) in 2 rnl of 10% sodium hydroxide solution at ooc was added 400 11-l (3.5 IIIOOl) of benzoylchloride. The mixture was stirred overnight to allow excess benzoylchloride to hydrolyze. The white solid which had formed was extracted into ether, dried (Na2004), and concentrated to dryness. This solid was chromatographed on silica gel using first dichloromethane followed by diethyl ether as eluting solvents. The product (Rf c.a •• 35, Et20/silica) was collected in 2 fractions which were free of impurities: M.P. 153-1550C; IR (CH2Cl2) 2980, 1690, 1515, 1490, 1395, 1370, 1200 cm-1; NMR (CilCl3) 5 7.8 (rn, 2H), 7.5 (rn, 3H) 6.95 (b, lH), 1.31 (s, 18H); Anal. Calcd. for C15H24N20: C, 72.54; H, 9.74; N, 11.28. Found: C, 72.40; H, 9.69; N, 11.11. 58 1,1,4,4-Tetra-~1-2-tetrazene To a 10 rnl (37). flask fitted with a balloon filled with oxygen was added 420 rng (2.9 mmol) of 1,1-di-~-butylhydrazine (24). Hydrazine 24 was stirred under oxygen at OOc for 2 hours as the clear oil became a white paste. The flask was then evacuated at room temperature to give a dry white solid. This solid was transferred to a vial and was evacuated in a vacuum desicator for 30 min. to give 156 mg (37%) of fine white needles which were sufficiently pure for elemental analysis: M.P. 49-SOOC; IR (mull) 2990, 1620, 1485, 1395, 1375, 1305, 1200, 1155, 930 cm-1; NMR (CDC13 )5 1.27 (s); UV (Et20), 230 nm (( 1500); Anal. Calcd. for C16H3 6N4: C, 67.55; H, 12.76; N, 19.69. Found: C, 67.29; H, 12.63; N, 19.44. Oxidatioo of 1,1-Di-.te.tt.-butylhydrazine (24) with .te.tt.Butyl.hypochl.orite. Preparation of Solutioos of 1,1-Diazene 16 for Visible Absorpt.ioo Spectroscopy. A 50 ml roundbottom flask was purged with argon and cooled to -127°C (n-propanol/liquid nitrogen bath). To this flask was added 5 ml of dimethyl ether, 150 mg (1.05 rrmole) of freshly prepped 1,1-di-~-butylhydrazine triethylamine. (24) and 134 mg (1.3 mmole) of To a second cooled flask (-1270C) was added 5 ml chlorotrifluoromethane {Freon-13), 0.5 rnl dimethyl ether and 115 mg {1.05 mmole) ~-butylhypochlorite. The pale yellow hypochlorite solution was added in small portions over 15 minutes to the rapidly stirred solution of hydrazine 24 causing a red color and white precipitate to form. The solution was stirred at -127°C for 45 minutes while a jacketed filter funnel with a 0.5 in. pad of celite 59 was cooled to -1270C. The red solution was quickly forced through teflon tubing into the cold filter funnel and a clear red solution was pulled through the frit into a flask at -1270C under vacuum. After stirring under vacuum to remove chlorotrifluorornethane, this solution was drawn into a low temperature uv-visible absorption cell which was maintained at -1250c to -1300C using liquid nitrogen. The visible spectrum was recorded at -1250C ± 1oc. (16) So1utioos for Proton Ym Spectroscopy. Tb prepare solutions of 1,1-diazene 16 for proton NMR spectroscopy, only 2 ml of dimethyl ether was used to dissolve the amines in the oxidation procedure described for visible spectroscopy. The red solution is then filtered into a cold flask (-1270C) containing 500 11.1 of deuteriodichlorornethane which had previously been passed through a short column of basic alumina. After removal of chlorotrifluoromethane under vacuum, dry argon was admitted to the flask and 0.3 m1 portions of the diazene solution were transferred, via 18 gauge teflon tubing under positive argon pressure, into 5 rnm NMR tubes (-1270C) which were equipped with serum caps. Proton spectra were recorded at -12ooc using a Bruker 500 MHz NMR spectrometer. flni..ssion Spectroscopy of 1,1-Dia.zene 16. The dimethyl ether:deuteriodichlorornethane (4:1) solutions of 1,1-diazene 16 in Srnn NMR tubes, prepared as described above for NMR spectroscopy, formed a clear glass when quickly frozen to -1960C. 60 Emission spectra were recorded using a non-comnercial spectrophotometer in Professor H.B. Gray's group at caltech. The 510 nm line of a 200 w mercury-xenon lamp was used for excitation. The emission spectrum was monitored through a Corning 3-67 filter with a Hamatsu R406 PM!'. Response factors for the R406 PM!' are 0. 719 ( 549 nm), 0.534 (600), 0.392 (651), 0.306 (699), 0.207 (750), 0.134 (801), 0.094 (849) and 0.066 (900). wanning the sample to -780C to decanpose 1,1-diazene 16 followed by refreezing the glass at -1960C causes complete loss of the emission signal. DeQJ *'{'OSition Kinetics for 1 ,1--Diazene 16. Decanposition of 1,1-diazene 16 was monitored by NMR spectroscopy. Acetone was added as an internal reference to solutions prepared for NMR spectroscopy as described above. The samples were placed in the probe of a Bruker 500 MHz spectrometer at the desired temperature and allowed to equilibrate for at least 5 minutes. The temperature measured by ·a thermocouple within the probe was calibrated at -920C using the chemical shifts of methanol. The actual temperature was found to be about 20C colder than the thermocouple indicated. Spectra were taken at regular time intervals by fourier transforming 32 pulses which were accumulated in 55 seconds. Time intervals and the number of spectra taken are summarized with the rate data in Table 8. Integrations were measured by cutting and weighing the diazene and acetone absorptions and by measuring the peak heights of these two singlets. Both methods gave the same rate, however rneasur ing peak heights was less sensitive to poor shimning and would typically give less scatter. All plots were first order with no 0.79 0 . 17 5 5 -102 -107 0.017 0.016 0.016 0.03 sa(xl03) 0.05 0.05 0.032 0.06 tsb(xl0 3 ) (b) Errorin k 1 at 95% confidence limit. 0 . 95 26 - 97 (a) Standard deviation of the slope. 1. 77 20 - ·94 k 1 ( sec-1 )( xl03) No . of Spectra for Decompositions of 16. Errors in First Order Rate Constants Temperature (OC) TABLE 8. (TI 1-' 62 evidence of curvature observable. Rates were determined by least squares analysis and are shown in Table 8 with error limits representing one standard deviation. Orldation of 1,1-Di~lhydrazine (24) with .t.e.t:t.- ButyJ.hypodll.orite. Procb:t Yields by Analytical VPC. To a chilled solution(-780C) of 179 mg (1.24 rmDle) of 1,1-di-.t,ell-butylhydrazine (24) and 126 mg (1.24 rmDl) of triethylamine in 5 ml of diethyl ether were added 66 mg of n-pentane and 70 mg of n-octane as internal standards. Hydrazine 24 was oxidized by slow addition of a cold solution (-7SOC) of one equivalent of .t,ell-butylhypochlorite in 5 ml diethyl ether. A white precipitate was formed during the addition but the red color of 1,1-diazene 16 was never observed. This mixture was stirred for 30 minutes and analyzed by analytical VPC (Dibutyl tetrachlorophthalate, 250C; carbowax 400, 250C; SE-54, 250C and 5.SOC). The hydrocarbon products, isobutane, isobutylene and hexamethylethane, were observed in 10%, 26% and 2% yields, respectively, for a canbined absolute yield of 3 8%. The yields are calculated as the number of mmoles of C4 groups for each product divided by twice the number of rmDles of 1,1-di-.tet.tbutylhydrazine (24) (based on C4 groups). hydrazine 24 was found to be 39%. The amount of unreacted '!be three hydrocarbon products were then formed in a corrt>ined yield of 62%, based on the amount of hydrazine 24 which had reacted. 'lWo additional products were observed • .tet.t-Butylchloride was found in 5% yield and had also been observed in NMR experiments. The second product was present in 63 approximately 0.3% yield and coinjected with authentic trans-1,2-di-~-butyldiazene. The assigrunent of this small conp:>nent as trans-1,2-di-~-butyldiazene cannot be made unambiguously, however. The yield and colwnn used for each product is sumnarized in Table 9. Orldatiat of 1,1-Di.-.teJ:t.--butyl.bydrazine (24) with Rickel Peroxide. Deteoainatiat of BydrocaJ:txn ProclJct Ratios by Analytical VPC. A solution containing 205 mg (1.42 mmole) of 1,1-di-~-butylhydrazine (24), 124 mg rroctane and 94.5 mg rrpentane in 10 ml of diethyl ether was prepared and cooled to -7a<>c. A ten-fold excess (3 gm, 13 rrmol available 0) of dry nickel peroxide49 was added to this solution through a solid addition funnel. The suspension was stirred for 90 minutes and then was filtered at -7a<>c using a jacketed frit. Analysis of the filtered solution by analytical VPC (SE-54, 250C, 550C; Dibutyltetrachlorophthalate, 250C) showed that hydrazine 24. had totally reacted. The only products observed fran the reaction were the 3 hydrocarbons, isobutane, isobutylene and hexamethylethane. They were formed quantitatively in the ratio 35:59:6 for isobutane:isobutylene:hexamethylethane. Significantly, none (<0.1%) of the rearranged product, 1,2-di-.tell-butyldiazene, was observed. Irradiatiat of 1,1,4,~1-rtetrazene (37) in a glass at-1960C. A solution of 68 mg tetrazene 37 and 40 Ill triethylamine in 1.1 ml of dimethyl ether:deuteriodichloromethane (8:3) was placed into 5 DBT, 25° DBT , 25° SE-54, 25° SE-54 , 55° Carbowax 400 , 250 Carbowax 400 , 25° Isobutane Hexamethylethane 1,1-Di-tert-butylhydrazine (24) tert-Butylchloride trans-1,2-Di-tertbutyldiazene Oxidation of Hydrazine 24. VPC Column (°C) Product Yields from tert-Butylhypochlorite. Isobutylene Compound TABLE 9 . 0.3 5 39 2 10 26 Yield (%) 0'1 .,. 65 nm NMR tubes and frozen at -1960C in a quartz finger dewar to form a clear glass. Irradiation of these samples with a 1000 w Xe arc lamp filtered through a visible absorbing-W transmitting filter (Corning 7-54) produced a red color which reached a steady intensity within 15 minutes. Gas bubbles are released when the glass is thawed at - 1270C but the red color persists until warmed to -780C. The red solution is quite viscous with a cloudy precipitate which prevents examination by The samples were renoved from the NMR tubes with partial Rtm.. decomposition and mixed with 2 m1 of chlorotrifluoromethane at -121oc. '!be red color remained in a thin layer floating on the freon with the precipitate. Filtration at -1270C passed a clear solution while the red color and precipitate remained on the frit. A similar red color was produced when a 2-methyl tetrahydrofuran glass containing tetrazene was used and when neat crystals of tetrazene 37 were irradiated at -1960C. A steady intensity for the red coloration was quickly reached in the 2-rnethyltetrahydrofuran glass. Warming the sample to -800C until colorless followed by refreezing and further illumination regenerated the red color. Di.(l-adamntyl)amne (30). A heavy walled glass tube was filled with 3.15 gm (0.015 mol) 1-adarnantylbranide and 4.5 gm (0.03 mol) 1-adarnantylamine and sealed under vacuum. After heating the tube at 2200C for 40 hours the solidified product was crushed and shaken in a separatory funnel containing 100 m1 of 20% NaOH solution and 250 m1 ethyl ether until the solid dissolved. When 10% aqueous HCl solution was added to the organic phase, di (1-adamantyl) amine hydrochloride precipitated as a 66 white solid. '!be product was filtered and shaken with 20% NaOH solution and diethyl ether to regenerate the secondary amine. Most of the excess mnoadamantylamine was washed into the aqueous HCl. solution as checked by TLC. The ether layer was dried (Na2004) and concentrated under reduced pressure to give a white solid which was further ~XIrified by column chromatography (neutral alumina/ether) to give 3.4 gm (80%) of di(l-adamantyl)amine: M.P. 196-1970C; IR (CCl4) 2930, 2860, 1490, 1455, 1355, 1310, 1165, 1100 cm-1; NMR (CDCl3) 8 2.0 (b, 6H), 1.78 (m, 12H), 1.64 (m, 12H). A 1 liter flask containing 11. Ogm (0. 039 ml) of di (1-adamantylamine (30) and 12.5 m1 dry pyridine stirred in 200ml dry diethyl ether was cooled with an ice/methanol bath to -200C. To this solution was slowly added 7.0 m1 of nitrosyl chloride which had been condensed into a graduated centrifuge tube, dissolved in cold ether, and transferred to a cooled jacketed addition fwmel. '!be reaction was allowed to warm to room temperature and stirred overnight. After adding the suspension to water, additional ether was needed to dissolve the product. The ether layer was washed with water and saturated NaHOOJ and dried (Na2004) • Removal of ether gave a yellow orange solid which smelled of pyridine. The product was redissolved in ether and passed through a column of neutral alumina which removed the orange color and the pyridine. The resulting yellow solid was recrystallized from ether/pentane and was ~XIre by TLC (ether/silica): M.P. 187-1880C; IR (CCl4) 2905, 2850, 1445, 1350, 1300, 1195, 1115, 1080, 970 cmrl; NMR (CDCl3) 8 2.38 (m, 12H), 2.14 (b, 6H), 1.73 (m, 67 12H)~ Anal. calcd. for C20H3oH20: C, 76.39~ H, 9.62~ N, 8.91. Found: C, 76.24~ H, 9.83~ N, 8.82. ~2,5-di~-butylpyrrole (33) • A solution of 23.78 gm (0.12 moles) 2,2,7,7-tetrarnethylocta3,6-dione4land 31.7 gm (0.24 moles) ~-butylcarbazate in 70 m1 glacial acetic acid was refluxed for 28 hours under nitrogen. After cooling, the reaction mixture was added to 100 m1 of water and the resulting suspension was extracted with 2xl50 m1 of ether. The ether layer was washed with SxlOO m1 10 N HCl, 3xl00 m1 saturated sodium bicarbonate, 100 m1 saturated sodium chloride and dried (Mg004) • Concentration of this solution gave a light yellow solid which was chromatographed (400 gm silica/methylene chloride) and 9.4 gm (40%) of N-aminopyrrole 33 was collected as the first elutant. '!be white solid (95% pure by analytical VPC, 4' Pennwalt-223, 2000C) was further purified by preparative VPC (5' Pennwalt 223, 1900C) or by recrystallization from hexane to g i ve fine white needles: M.P. 66-67°C~ IR (CC1 4 ) 3380, 3300, 3103, 2960, 1476, 1460, 1389, 1360, 1250, 950, 870 cm-1~ NMR (CDCl3) 8 5.38 (s, 2H), 4.23 (b, 2H), 1.35 (s, 18H)~ mass spectrum, m(z (rel. intensity) 194 (14), 179 (19), 137 (3), 123 (25), 57 (15)~ Anal. Calcd. for C12H22N2: 11.41~ N, 14.42. Found: C, 74.17~ H, C, 74.21~ H, 11.49~ N, 14.31. 3 ,6-Di~-but:ylpyrida.zine (36) • A solution of 1.5 gm (7.6 mmoles) 2,2,7,7-tetramethylocta3,6-dione (34) and 0.49 gm (15.2 nrnoles) hydrazine in 3.5 ml glacial acetic acid was refluxed under nitrogen for two hours. The mixture 68 was exx>led, made basic with a saturated solution of potassium carbonate and extracted 3x25 m1 with ethyl ether. '!be ether extracts were washed with 10 ml of water, dried (Mg004) and concentrated to a white solid. Recrystallization fran ether/pentane gave 0.55 gm of fine white needles (35%): M.P. 170-1750C [lit. M.P. 175-1770C] ;50 IR (Nujol) 3035, 1585, 1475, 1428, 1360, 1148, 865 cm-1; NMR (CDCl3) 8 7.40 (s, 2H), 1.44 (s, 18H). l,l'Azo-2,5-di-~-butylpfrrole (37). Using a solid addition funnel, 460 mg (1.1 equiv.) of lead tetraacetate was slowly added to a solution of 200 mg (1.0 rnmole) of ~amincr2,5-di-~-butylpyrrole cooled to OOC under nitrogen. then reddish brown. (33) in 10 rn1 dry methylene chloride '!be solution first turned yellow and After an hour of stirring at ooc the reaction mixture was washed with water, saturated sodium bicarbonate, dried (Mg004) and concentrated. The reddish residue was chromatographed (silica/methylene chloride) to give 50 mg (25%) of light yellow tetrazene 37: M.P. 102-1030C; IR (CH2Cl2) 3105, 2990, 1460, 1397, 1366, 1200, 1100, 1018 cm-1; NMR (CDCl3) o 6.0 (s, 4H), 1.30 (s, 36H); exact mass calcd. for C24H4oN4 384.325, found 384.326. lr(2,5-Di~l-R'-pyrrolyl)dimet:hylsulfnxjmjne (38). Using a solid addition funnel, 0.34 gm (0.77 mmole) lead tetraacetate was added over an hour to a solution of 150 mg (0. 77 mmol) of ~amincr2,5-di-~-butyl pyrrole (33) in 4 rn1 each of methylene chloride and di.nethyl sulfoxide at ooc. After stirring for an additional one and one-half hours the mixture was taken up in more 69 methylene chloride and washed with water, saturated sodium bicarbonate and saturated sodium chloride. After drying (Mg004) the solvent was removed to give a reddish solid which was recrystallized (chloroform/hexane) to give 70 m; (34%) of near white crystals: M.P. lll-ll20C (decamp.)J IR (CB2Cl2) 3110, 2985, 1396, 1363, 1307, 1205, 1058, 1023, 944 cm-lJ NMR (CDC13) o 5.80 (s, 2H), 3.03 (s, 6H), 1.36 (s, 18H)J Anal. Calcd. for C14H46N20S: Found: C, 62.18; H, 9.69; N, 10.36. C. 62.18; H, 9.89; N, 10.28. tr(2,5-Di~l-R'-pyrrolyl)-d(;-dimetbyl.sulforjmjne (3~) • The deuterated dimethylsulfoxide adduct was prepared using dimethylsulfoxide-d6 in the same procedure used to make 38: IR (CH2Cl2) 3105, 2970, 2260, 2140, 1393, 1361, 1210, 1058, 1020, 959 cm-1; NMR (CDCl3) o 5.80 (s, 2H), 1.38 (s, 18H). Solution 'Dlen.>lysis o£ tr(2,5-Di~-but.yl-R'-pyrrolyl) A pyrolysis tube containing 300 p.l of a 0. 059 .H tetrahydrofuran solution of sulfoximine 38 was sealed under vacuum and placed in an oil bath at 500C. After 4 hours the tube was opened and the 'mF was reooved under reduced pressure. Analysis of the products by NMR showed that 15% of the dimethylsulfoxide adduct had decomposed cleanly to give tetrazene 37 and free dimethylsulfoxide. Solution 'Dlermolysis o£ tr (2,5-Di.-:tert-but.yl-N'-pyrrolyl) -i'Pdimetbylsulforimine (38-d6) in the Presence o£ Dimethyl.sul.foride. A solution of 5.06 mg (0.0184 mmoles) deuterated adduct (3s-a6) 70 and 4.3 mg (3 equivalents) dimethylsulfoxide in 0.5 ml of carbon tetrachloride was placed in a 5 mn NMR tube, degassed and sealed under vacuum. The Rom spectrum of this solution shows that the dimethylsulfoxide adduct is completely deuterated before thermolysis. After heating the tubes at 500C for 3-1/2 hours the dimethylsulfoxide adduct was 33% deuterated at the sulfoximine methyls which corresponds to 89% equilibration of d6-dimethylsulfoxide between the dimethylsulfoxide adduct and free dimethylsulfoxide. After 12-1/2 hours of thermolysis dimethyl sulfoxide was 97% equilibrated. Although the solution had becane slightly brown, tetrazene 37 or any other product were not ooserved in the NMR spectrum. Atteapt.ed Flash Pyrolysis of Dimetbylsulfoximi ne (38) • Dimethylsulfoximine 38 was placed in a sublimation tube near the inlet of a quartz tube heated to 4500C with a Hodgkin's furnace. A liquid nitrogen cooled finger was placed at the outlet of the heated tube and the system was· evacuated to 0.2 torr. Sublimation did not occur until the sublimation tube was heated to 90-lOOOC and the white crystals began to yellow. '!he yellow condensate on the cold finger did not change color when thawed and NMR analysis showed this material to be composed of dimethylsulfoxide and 2,5-di-~-butylpyrrole (35). The residue in the sublimation tube contained both tetrazene 37 and pyrrole 35 but neither dirnethylsulfoximine 38 or dimethylsulfoxide were present. Dropping of a solution of dirnethylsulfoximine 38 directly into the pyrolysis tube was also attempted. A solution of 15.7 gm of the dimethylsulfoximine in 25 m1 of decalin was added dropwise into a 71 vertical quartz tube heated at 3600C Wlder vacut.nn. A golden brown solid was condensed at the bottan of the hot tube in a 50 m1 flask at -1960C. After removal of rost of the decalin under vacut.nn at 340C the residue was foWld to contain 82% unreacted dimethylsulfoximine 38 and 18% tetrazene 37. 72 References (1) For reviews of l,l~iazene behavior see: (a) Lemal, D.M. In "Nitrenes"; Lwowski, w., Ed.; Interscience:New York, 1970; Chapter 10. (b) Ioffe, B.V.; Kuznetsov, M.A. (Engl. Trans!. l 1212., 41, 131. Russ. Chern· Rev· (c) Lwowski, W. In "Reactive Intermediates"; Jones, M. and Moss, R., Eds.; Wiley, New York, 1978; Olapter 6. (2) For a review of 1,2-diazene behavior see: Engel, P.S. Chern· ~~' 80, 99. {3) (a) Lathan, W.A.; CUrtiss, L.A.; Hehre, W.J.; Lisle, J.B.; Pople, J.A. Prog. pWs. Org. Chern• .l21.i, ll, 175. (b) Ahlrichs, R.; Staemmler, v. Chern. pWs. Lett. ~' 37, 77. (c) Wagniere, G. Tbeor. Olim. Act.a~, 31, 269. (d) Baird, N.C.; Wernette, D.A. Can. J. Chern. ill:]_, 55, 350. (e) Davis, J.H.; Goddard, W.A. J. Am· Chern. Soc. ill:]_, 99, 711. (f) Pasto, D.J.; Chipnan, D.M • .ibid, .J.21i, 101, 2290. (g) casewit, C.J.; Goddard, w. A• .ibid .lliDl, 102, 4057. (4) (a) Hinsberg, W.D.; Dervan, P.B. J. Am· Chern. Soc • .l.21.a, 100, 1608. (b) Hinsberg, W.D.; Dervan, P.B • .ibid, .J.21i, 101, 6142. (5) Schultz, P.G.; Dervan, P.B. J. Am· OJem. Soc. ,lliD2., 102 878, (6) Hinsberg, W.D.; Schultz, P.G.; Dervan, P.B. J. Am· Chen· Soc. ll!l2,, 104, 766. (7) Schultz, P.G.; Dervan, P.B. J. Am· Chan- Soc. In press. 73 (8) Schultz, P.G.; Dervan, P.B. J. Am· Chern· Soc. J.2!U, 103, 1563. (9) Hinsberg, W.D., Ph.D. 'lbesis, California Institute of Tecmology, Pasadena, california, 1979. (10) Dervan, P.B. 7 Squillacote, M.E.; Lahti, P.M.; Sylwester, A.P.; Roberts, J.D. J. Am· Chern· Soc • .1.2.!U., 103, 1120. (11) Golzke, v.; Groeger, F.; <l:>erlinner, A.; Ruchardt, C. Nouv. J. Chim. (12) ma, 2, 169. (a) Chae, W.; Baughnan, S.A.; Engel, P.A.; Bruch, M.; Ozmeral, C.; Szilagyi, S.; Tirrberlake, J.W. J. Arne Chern. ~lial, 103, 4824. (b) Engel, P.S.; Melaugh, R.A.; Page, M. A.; Szilagyi, s., Ti.nt>erlake, J.W. ibid l216_, 98, 1971. (13) Schulz, A.; Ruchardt, c. Tett, Lett. l216_, 43, 3883. (14) Prochazka, M. Collect. Czech, Chem- Cqrmun, ..1212., 41, 1557. (15) (a) Applequist, D.E.; Klug, J.H. J. Org. Chern. ~' 43, 1729. (b) I.uh, T,; Stock, L.M • .ibid~' 43, 3271. (c) Fort, R.C.; Hiti, J. ibid~' 42, 3968. (16) Firestone, R.A. J. Org, Chern· ~, 45, 3604. (17) March, J. "Advanced Organic Chemistry", 2nd ed.; McGraw-Hill:New York, 1977; p.45. (18) Lemal, D.M.; Rave, T.W.; McGregor, S.D. J, Arne Cherne Soc, ~, 85, 1944. (19) Back, T.G.; Barton, D.H.R. J. Chern, Soc. Perkin I ~' 924. (20) C, H and N analyses were obtained for the benzamide derivative because of the lability of hydraziDe 24, see experimental section. 74 (21) (a) Cosse-Barbi, A.; Dubois, J.E. ~rochimica Acta ill2,, 28A., 523. (b) Cosse-Barbi, A.; Dubois, J. E• .ibid ill2,, 28A., 561. (22) Tbe infrared spectrum of diazene 16 is unattainable due to solvent limitations at -1250C, but the N-N stretch for 16 is expected to be similar to stretches observed for diazenes 3 and 4 at 1595 cm-1 and 1638 cm-1, respectively.6 ( 23) We thank. William Croa.smun for assistance on the 500 MHz NMR experiments at low temperature. (24) Van Geet, A.L. Anal. Chen· ~' 42, 679. (25) Tbe precisian for temperature control in the probe appears to be quite good during a single kinetic run, however temperature calibration for accuracy can vary by as much as a degree on different days. (26) ~G* values were calculated to be 12.55, 12.55, 12.25, and 12.39 kcal/mole at 1790, 1760, 1710, and 1660K, respectively. (27) The errors are reported as the 95% confidence limits instead of single standard deviations because of the small number of points in the analysis. (28) Using the preexponential term of 1013 typical of 1,1-diazenes 3 and 4 ( 13 •7 and 12. 4 , respectively6) , an Ea of about 13 kcal/mol is calculated for the decanposition of 1,1-diazene 16. (29) Ermer, 0.; Lifsan, s. Tetrahedron 121!, 30, 2425. (3 0) Mill, T. ; Stringham, R. S. Tet. Lett. .l..9.6.2,, 23, 1853 • (31) Tanner, D.D.; Rahimi, P.M. J. Am· Chen. Soc. ~' 104, 225. 75 (32) Martin, G.7 Maccoll, A. J. Chern· Soc. Perkin II lli:L, 1887. (33) Duan, D.C.7 Dervan, P.B., In preparation. (34) Tsuji,T.7 Kosower, E.M. J. Am· Chern· Soc • .J.21l, 93, 1992. (35) Malatesta, V.7 Ingold, K.U. J. Am· Chern. Soc. ll!ll., 103, 3094. J. Am· Chen· Soc. J.2ll, 95, 3228. (36) Roberts, J.R.7 Ingold, K.U. (37) z.k>rat, C.7 Rassat, A. Bull. Soc. Chim. Fr • .J.21l, 891. (38) l-Iodo-4-methoxybicyclo[2.2.2]octane can be made from 1,4-dimethoxybicyclo[2.2.2]octane and HI by heating them in a sealed tube at 1700C, George Geller private carmunication. (39) This reaction has been used to prepare 1-camphenyl-tert-butylnitroxide from 1-nitrocamphene and tert-butyl grignard. Brunei, Y.7 Lemaire, H.; Rassat, A. Bull. Soc. Chim. Fr. ~, 1895. (40) Ramasseull, R.; Rassat, A. Bull. Soc. Chim. Fr. ~, 3136. (41) Ito, Y.; Konoike, T.; Harada, T.; Saegusa, T. J. Am· Chern· ~ ~, 99, 1487. (42) Koga, G.7 Anselme, J.P. J. Org. Chern· ill.Q, 35, 960. (43) Bock, H.; Kaapa, K.L. Angew. Chen. Internat. Ed • .1222., 1, 264. (44) The cell was designed by William Hinsberg and detailed drawings can be found in his thesis.9 (45) Air Products and Chemicals, Inc., Advanced Products Department, P.O. Box 538, Allentown, PA 18105. (46) Amersil, Inc., 685 Ramsey Avenue, Hillside, NJ 07205. (47) Alpha Wire Corp., 711 Lidgerwood Avenue, Elizabeth, NJ 07207. 76 (48) Kornbllml, N.; Clutter , R.J.; Jones, W.J. J. Am· Chern. Soc. ~, (49) 78, 4003, Nakagawa, K.; Konaka, R.; Nakata, T. J. Org. 1597. Chern· ~' XI,