The Thermolysis of Cyclopropyl Azides.

ments (cf. Ref.[51)and lactonization experimentsf6! The
synthetic ( F )-threo-cis-methylester ( 8 ) [m.p. (hydrochloride) 176.4-1 77 "C] was spectroscopically and chromatographically identical with the (-)-methyl ester [m.p.
(hydrochloride) 185"C] obtained by degradation of the
alkaloid.
membered ring protons in carbon tetrachloride at 60 MHz:
triplets (each 2 protons) at r=6.75 and 6.15, J=3.5 Hz;
IR: C=N at 1568 cm-']. Reduction by LiAIH, converts
H
Analogous reactions with trans-5-octenal (3) have led to
synthesis of the erythro-cis- and the erythro-trans-compound (11 a ) ; these stereoisomers differ significantly from
(8) in their NMR and IR spectra.
The (2S,6R,I'S)-configurationof ( - ) - ( 8 ) follows unambiguously from the chiroptical data for ( - ) - ( 8 ) and (-)-cis(11b) ['] together with the threo-configuration that has
now been proved. Thus palustrine has structure (10).
Received: September 14,1972 [Z 755 IE]
German version: Angew. Chem. 85,172 (1973)
H
H
H
R'
R'
( 1 ) - (3)
Yield [ %]
~ ~ 7
Compound
( I ) , R ' = R 3 = H , R 2 =C,H,
(2),R'=C6H5,R2=R3=H
f'3),R1=H,R2+R3=-(CH2)4-[8]
36
40
21
B.p. ["C/torr]
1
45-50/0.003
4548/0.001
35-38/0.01
[l] W Dietsche and C . H . Eugster, Chimia 14, 353 (1960).
[2] C . Mayer, W Trueb, J . Wilson, and C . H . Eugster, Helv. Chim. Acta
51, 661 (1968).
[3] G. Eglinton, E. R. H . Jones, and M . C. Whiting, J. Chem. SOC.1952,
2873.
[4] J . C. Collins, Tetrahedron Lett. 1968, 3363.
[5] J . B. Lambert, R G. Keske, R. E. Carhart, and A . P . Jocanovich,
J.Amer. Chem. SOC.89,3761 (1967); E. Fujita and K . Fuji, J.Chem. SOC.
C 1971, 1651.
[6] C. L. Green, unpublished work, 1968.
[7] G . Mukherjee, Diploma thesis, Universitat Zurich 1970
( 4 ) into 2-phenylazetidine (.5)['],
and addition of butyllithium into 2-butyl-2-phenylazetidine (6) ['I ;additionally,
thermolysis of (2) affords 21 % of benzonitrile and ethylene,
of which the latter was identified as 1,2-dibromoethane.
o<H
The Thermolysis of Cyclopropyl Azides"]
By Giinter Szeirnies, Udo Siefken, and Rolf Rinck"]
Cyclopropyl azides were unknown until very recently
when some representatives of this class of compounds were
synthesized by special
In the nitrogen-transfer reaction from p-toluenesulfonyl
azides to anions of cycIopropylamines[5~we have found a
more general method of preparing cyclopropyl azides.
N C&
H
(5)
c)<('CHZ)~-CH~
N
H
C6H5
( 7)
(6)
Nitrogen cleavage from (1) in o-dichlorobenzene at 110°C
gave, together with styrene (57%) and hydrogen cyanide, a
rapidly polymerizing oil that we have not yet been able to
purify. Thermolysis of (3) in carbon tetrachloride in a
Table I.
Rate constants of the azide thermolysis in o-dichlorobenzene [9].
Concn.
[molPl
Temo.
["CIS
k. x
(1)
0.0470
0.0470
0.0470
0.0476
0.460
0.0476
0.0460
0.0460
0.0460
112.8
126.1
135.2
102.5
102.5
112.7
3.28
12.3
29.2
3.12
3.20
8.70
100.0
112.7
120.0
150.0
165.0
175.0
2.31
8.10
16.4
1.04
4.25
10.0
(2)
(31
(7)
0.0562
0.0562
0.0562
Thermal removal of nitrogen from the cyclopropyl azides
( l ) - ( 3 ) can be effected between 100 and 120°C. For
example, after 2 h in boiling toluene, (2) yielded 76% of
2-phenyl-l-azetinef71( 4 ) lINMR signals of the four____
[*] Dr. G . Szeimies, U. Sieken, and R. Rinck
Organisch-chemisches Laboratorium der Universitat
8 Munchen 2. Karlstrasse 23 (Germany)
A n g e w Chmi. internat.
lo4
Azide
Edit. 1 Vol. 12 ( l Y 7 3 )
No. 2
[6- '1
E,
[k^cal/mol]
AH*
[kcal/mol]
ASi
[cal molden-'
30.5
29.7
2.0
28.9
28.2
0.0
28.6
27.8
- 0.6
34.2
33.3
1.1
sealed tube at 120°C led to 90% of cyclohexene and a
small amount of a resin.
It is noteworthy that the thermolysis temperature of the
cyclopropyl azides is much lower than that of alkyl azides.
Table 1 gives the first-order rate constants and the activation parameters of the thermal decomposition of the
azides ( I ) , ( 2 ) , and (3) and of I-phenylcyclobutyl azide (7)161
An activation energy of 38.5 kcal/mol['*] is required for
removal of nitrogen from isopropyl azide, and of 47.5
kcal/mol for that from cyclohexyl azide" 'I. What makes
cyclopropyl azides so labile thermally? Loss of the threemembered ring strain cannot explain this great acceleration of the removal of nitrogen from three-membered ring
azides : assuming comparable transition states, the fourmembered ring azide (7) should decompose faster than
the corresponding three-membered ring azide (2) if loss
of ring strain has a decisive effect on the energy of the activation
since the cyclopropane and cydobutane
rings have approximately the same ring strains of 27.1
kcal/mol and 26.2 kcal/mol, re~pectively['~],
whilst the
azetine system which is the main thermolysis product of the
three-membered ring azide should be considerably more
strained than the I-pyrroline system which is the thermolysis product of the four-membered ring azide. The activation
enthalpy of thermolyis of the cyclobutyl azide (7) is about
5 kcal/mol greater than that of the cyclopropyl azide (2),
corresponding to a rate ratio of 1:525 at 102.5"C, and this
indicates that the cyclopropane ring is not broken to an
appreciable extent in the rate-determining step of nitrogen
elimination. Removal of nitrogen from the cyclopropyl
azide should thus not be wholly synchronous with ring
expansion and cheletropic elimination['41, but the reactions occur one after the other.
With this in mind we see a possible explanation of the
considerable facilitation of removal of nitrogen from
cyclopropyl azides by the electron donor effect of the
cyclopropane system to the empty p orbital of the nitrene
nitrogen in the orbital combination represented in (8),
the transition state being able to profit therefrom: the
occupied asymmetric Walsh orbital['51of the cyclopropane
can fill the electron deficiency on the nitrene nitrogen
caused by nitrogen elimination[161.
This "vertical stabilization"[171of the nitrene nitrogen by
the cyclopropyl group has an analogy in the stabilization
of a carbenium ion center by a cyclopropyl substituent.
This interaction can also be recognized in the increased rate
of solvolysis of certain cyclopropylmethyl compounds" :
acceleration of the solvolysis and of the "nitrogenolysis"
on cyclopropyl substitution should thus have similar
origins. The formation of the end products can be easily
explained by the interaction represented in (8): charge
transfer from the three-membered ring to the nitrene
nitrogen weakens the bonds C'-C2 or C'-C3 in the
cyclopropane and thus facilitates the rearrangement to the
azetine or the cheletropic elimination of nitrile to give
ethylene" 'I.
That the two last-mentioned compounds are not products
of a subsequent thermal decomposition of the I-azetine
system can be demonstrated conclusively for pyrolysis of
(2) : benzonitrile is formed at 102.5"C in a first-order
162
reaction with k,=2.91 x
[s-'1, and this value is
identical within experimental error with k , of the azide
t hermolysis.
Received: October 27,1972 [Z 756 IE]
revised : December 6,1972
German version : Angew. Chem. 85,173 (1973)
[1] This work was supported by the Deutscbe Forschungsgemeinschaft.
[2] W Kirmse and H . Schutte, Chem. Ber. 101, 1674 (1968); J. Amer.
Chem. SOC.89,1284 (1967).
[3] A. 8. Levy and A . Hassner, J. Amer. Chem. SOC.93, 2051 (1971).
[4] J . E. Galle and A . Hassner, J. Amer. Chem. SOC.94, 3930 (1972).
[5] W Fischer and J.-P. Anselme, J. Amer. Chem. SOC.89,5284 (1967);
7: R . Steinheimer, D. S . Wulfman, and L. M . M c Cullagh, Synthesis 1971,
325.
[6] Yields refer to the amine that reacted. Recoveries were 41 y4 of the
amine used for ( I ) , 20% for (2). and 65% for (3).
171 The products obtained gave correct values on elemental analysis.
IR, NMR, and mass spectra are in agreement with the structures
proposed.
[8] This compound was recently prepared by a similar method by
D.S. Wuljman and 7: R . Steinheimer, Tetrahedron Lett. 1972,3933.
[9] The rate constants were determined by IR spectrophotometry
with the asymmetric stretching vibration of the azide system as reference band.
[lo] G. Geiseler and W Konig, Z . Phys. Chem. (Leipzig) 227, 81 (1964).
[ I l l P. Walker and W A. Waters, J. Chem. SOC.1962, 1632.
1121 The main product of thermolysis of (7) is 2-phenyl-I-pyrroline.
[13] R . B. lhrner, P . Goebel, B. J . Mallon, W u. E. Doering, J . F. Coburn,
and M . Pomerantz, J. Amer. Chem. SOC.90, 4315 (1968). and literature
cited therein.
[14] R. B. Woodward and R . Hofmann, Angew. Chem. 81, 797 (1969);
Angew. Chem. internat. Edit. 8, 781 (1969).
[t5] A . D. Walsh, Trans. Faraday SOC.45, 179 (1949); R. Hqffmann,
Tetrahedron Lett. 1970, 2907; H . Giinther, ibid. 1970, 5173; P. Bischol;
R. Gleiter, E. Heilbronner, K Hornung, and G. Schroder, Helv. Chim.
Acta 53, 1645 (1970).
[is] It is not necessary to formulate free cyclopropylnitrene as an
intermediate in the decomposition of the cyclopropyl azide. A strongly
asynchronous process with extensive nitrogen-nitrogen bond cleavage
in the transition state and late occurrence of the opening of the bond
in the cyclopropaneskeleton would profit from thesame interaction (8).
[17] 7: G. Taylor, W Hanstein, H . J . Berwin, N . A. Clinton, and R . S .
Brown, J. Amer. Chem. SOC. 93, 5715 (1971).
[18] Z . Majerski and P . uon R . Schleyer, J, Amer. Chem. SOC.93, 665
(1971) and literature cited therein. The difference in the activation
enthalpies for solvolysis of cyclobutylmethyl tosylate [19] and cyclopropylmethyl tosylate [20] amounts to 8 kcal/mol.
[I91 K . B. Wiberg and B. A . Hess Jr., J. Amer. Chem. SOC.88, 4433
(1966).
1201 D. D. Roberts, J. Org. Chem. 29, 294 (1964)
[21] Attention is drawn to the analogy with decomposition of cyclopropyldiazomethanes. For this see W Kirmse: Carbene Chemistry.
2nd Edit., Academic Press, New York 1971, pp. 467 ff.
A Direct Cyclobutene-Bicyclobutane Valence
Isornerization''II**l
By Giinther Maier and Manfred Schneider"'
Irradiation of tricycle ( I ) leads to fission into tetramethylcyclobutadiene and tetramethylpyridazinet'l. If it were
possible to prepare the heterocycle (2) that is valenceisomeric with ( I ) , a substance would be available whose
photofragmentation might open a route to tetramethyltetrahedrane.
[*I
Prof. Dr. G. Maier and DipLChem. M. Schneider
Fachbereich Chemie der Universitat
355 Marburg, Lahnberge (Germany)
[**I This work was supported by the Deutsche Forschungsgemeinschaft and the Fonds der Chemischen Industrie.
Anqew. Chem. internat. Edit.
/ Vol. I 2 (1973) 1 No. 2