Azulene–Naphthalene-Type Rearrangements in Benz[a]azulene and Cyclohepta[b]indole

Article abstract of DOI:10.1002/chem.201603389

Flash vacuum pyrolysis (FVP) of benz[a]azulene yields phenanthrene and 2-ethynylbiphenyl. FVP of cyclohepta[b]indole similarly yields phenanthridine and 2-cyanobiphenyl. The reversibility of the reactions is demonstrated by FVP of 2-ethynylbiphenyl and 2-isocyanobiphenyl. All the observed reactions are in accord with the norcaradiene–vinylidene mechanism of the azulene–naphthalene rearrangement, whereas other proposed mechanisms are ruled out.

Full text of DOI:10.1002/chem.201603389

DOI: 10.1002/chem.201603389
Communication
&
Reaction Mechanisms
Azulene–Naphthalene-Type Rearrangements in Benz[a]azulene
and Cyclohepta[b]indole
Curt Wentrup*[a] and Jꢀrgen Becker[b]
ysis of 4–13C-azulene at 11808C/10ꢀ3 hPa (Scheme 1);[7] and 2)
Abstract: Flash vacuum pyrolysis (FVP) of benz[a]azulene
the 9,10-diradical mechanism of Scott and Kirms based on py-
yields phenanthrene and 2-ethynylbiphenyl. FVP of cyclo-
rolysis of 1–13C-azulene in an N2 flow system at 700–9008C
hepta[b]indole similarly yields phenanthridine and 2-cya-
with contact times of 0.3–3 min (Scheme 2).[8,9] A previously
nobiphenyl. The reversibility of the reactions is demon-
proposed pericyclic bicyclobutane mechanism[10] has been dis-
strated by FVP of 2-ethynylbiphenyl and 2-isocyanobi-
counted because of the high calculated activation barri-
phenyl. All the observed reactions are in accord with the
er,[5,6,11,12] and because it cannot account for most labeling re-
norcaradiene–vinylidene mechanism of the azulene–naph-
sults.[13]
thalene rearrangement, whereas other proposed mecha-
In addition, two intermolecular mechanisms involving the
nisms are ruled out.
addition of hydrogen atoms (or methyl radicals)—the methyl-
ene walk and spiran pathways—(Scheme 3) have been put for-
ward and promoted strongly by Alder and co-workers,[12] but
Introduction
suffer from the fact that none of the numerous postulated in-
termediates have ever been observed. The methylene walk
process allows the migration of a CH2 group from the seven-
to the five-membered ring (9!10). In the spiran route,
a carbon migration in the spirocyclic intermediate, 11!12,
and a 1,5-H shift 13!14 are required to turn a 2-subsituted
azulene into a 1-substituted naphthalene.
The thermal isomerization of azulene to naphthalene (A–N re-
arrangement) [Eq. (1)] has intrigued organic, physical and com-
putational chemists for nearly 70 years.[1] Although the reaction
is usually achieved by high-temperature flash, flow, or static
pyrolysis, recent variations include the graphite-sensitized, mi-
crowave-induced rearrangement[2] and a room temperature
cationic reaction promoted by BF3·Et2O.[3]
The mechanisms in Scheme 3 were developed largely to ex-
plain the rerrangements of methyl-substituted azulenes, for ex-
ample the fact that 2-[13C-methyl]-2–13C-azulene affords both
2-[13C-methyl]-2–13C-naphthalene and 1-[13C-methyl]-1–13C-azu-
lene (sealed tube, 4408C, 4 h),[12,14] but they ignore the fact
that azulene itself undergoes skeletal rearrangements under
these reaction conditions as summarized in Scheme 4.[15] For
example, 1-phenyl-1–13C-azulene isomerizes to 2-phenyl-2–13C-
azulene, and 1-phenyl-3–13C-azulene isomerizes to 2-phenyl-3a-
13C-azulene, whereas labels and substituents in the seven-
membered ring remain in their original positions.
The best available activation energy for the thermal A–N re-
arrangement, determined in shock tube experiments at 1000–
16008C with azulene highly diluted in Ar, is 63 kcalmolꢀ1 with
It has to be realized that no single mechanism for the A–N
rearrangement can explain all the experimental results,[7,16]
which is not surprising for a reaction taking place at tempera-
tures of 10008C or higher. Moreover, because of these extreme
conditions, authors often use poorer vacua (10ꢀ1–10ꢀ2 hPa) or
pack the pyrolysis tubes with quartz chips to increase the con-
tact times to seconds, or use higher pressure flow conditions
with contact times of minutes to lower the reaction tempera-
ture. Worse still, the reactions may be performed in sealed
tubes at for example 4008C for several hours.[12,14] It is no sur-
prise that multiple reactions, including radical-induced ones,
may take place under such conditions, and it is quite plausible
that the A–N rearrangement can be catalyzed by hydrogen
atom addition and elimination. We have shown that, under
FVP conditions with contact times in the millisecond range,
about 24% of intermolecular reaction occurs, as indicated by
the formation of [D0]naphthalene, [D1]naphthalene, and
an Arrhenius pre-exponential factor of 1012.93 [4]
.
Importantly, the
shock-tube reaction was found to be cleanly unimolecular.[4a]
A
somewhat higher activation energy of 73 kcalmolꢀ1 is predict-
ed by G3(MP2,CC) calculations,[5] whereas DFT calculations pre-
dict activation energies of 80 kcalmolꢀ1 or more.[6]
Two intramolecular mechanisms for the A–N rearrangement
have been proposed, 1) the norcaradiene–vinylidene mecha-
nism by Becker, Wentrup, Katz and Zeller based on flash pyrol-
[a] Prof. Dr. C. Wentrup
School of Chemistry and Molecular Biosciences
The University of Queensland, Brisbane, Qld 4072 (Australia)
E-mail: wentrup@uq.edu.au
Homepage: http://researchers.uq.edu.au/researcher/3606
[b] Dr. J. Becker
Fachbereich Chemie der Philipps-Universitꢀt
D-35037 Marburg (Germany)
Chem. Eur. J. 2016, 22, 1 – 6
1
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
Scheme 1. The norcaradiene-vinylidene mechanism (Becker, Wentrup, Katz, Zeller).
Scheme 2. The diradical mechanism (Scott and Kirms).
Scheme 4. Azulene–azulene rearrangement.
[D2]naphthalene, by FVP of [D]azulene-4 at 11808C/10ꢀ3 hPa,
whereas no change took place in the azulene itself.[7] However,
this does not say anything about the mechanism of the A–N
rearrangement, as it is necessary to consider the effect of
chemical activation[17] in the FVP experiments. This will be
more than 100 kcalmolꢀ1 in the naphthalene formed by rear-
rangement of azulene (the exothermicity plus the activation
energy). H atom loss takes place from highly excited naphtha-
lene after its formation by rearrangement of azulene, and the
dary rearrangements of naphthalenes have to be considered in
all low-pressure pyrolysis investigations of the A–N rearrange-
ment. The shock-tube experiments[4a] clearly indicate unimole-
cularity of the rearrangement. Moreover, the rearrangement
also occurs in highly vibrationally excited ground state azulene
under wall-less conditions using one-, two- or three-photon ex-
citation.[5,18–21] It is clear, therefore, that unimolecular reaction
mechanisms need to be considered, and to obtain mechanisti-
cally meaningful results, the experiments should be conducted
under the best possible FVP conditions, that is, high vacuum
and short contact times. In our search for reactions that allow
distinctions to be made between the several proposed mecha-
nisms we are pleased to report two examples herein.
barrier is in the range 71–80 kcalmolꢀ1 [5]
, well within the chem-
ical activation range and comparable to the barrier of the A–N
rearrangement itself. Even the barrier for the C1,C2-automeri-
zation in naphthalene, which has been given as around 86 kcal
molꢀ1 [16]
is within the chemical activation range. Hence, secon-
,
Scheme 3. The H-addition/methylene walk and spiran pathways (Alder et al.)
Chem. Eur. J. 2016, 22, 1 – 6 www.chemeurj.org
&
&
2
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
The synthesis of benz[a]azulene 15 by ring expanding rear-
The reaction mechanism presented in Scheme 6 leads initial-
rangement of 2-fluorenylcarbene was described previously.[22] ly to 2-isocyanobiphenyl 24. Since isocyanides rearrange clean-
The formation of 15 together with phenanthrene 19 by FVP of ly to nitriles with activation energies of the order of 35 kcal
2-ethynylbiphenyl 18 is well-known,[23] and the isoxazolones 20 molꢀ1 [27]
they are always difficult to detect under FVP condi-
,
and 21 provide a useful entry point (Scheme 3).[23c] There is tions. Nonetheless, when the pyrolysis of cyclohepta[b]indole
considerable experimental and computational evidence for the 22 was repeated at 4008C/10ꢀ5 hPa, very little rearrangement
interconversion of 18 and the vinylidene 17.[23] A pyrolysis of occurred, but 2-isocyanobiphenyl 24 was detectable by its
benz[a]azulene under static conditions (4408C, 2 h) reportedly characteristic IR absorption at 2120 cmꢀ1, which was identical
yielded both phenanthrene and anthracene.[24] This outcome with that of an authentic sample prepared by a literature
was not explained, but, as elaborated above, the reaction con- method.[28]
ditions are far from ideal for the study of unimolecu-
lar rearrangements, and it is quite possible that free
radical processes intervene.
We subjected 15 to FVP in the temperature range
750–12008C and found that phenanthrene 19 and 2-
ethynylbiphenyl 18 are obtained as the only signifi-
cant products as determined by gas chromatogra-
phy; at 12008C the yields of 19 and 18 were 95 and
2%, respectively. Anthracene was not detected. The
results of FVP of 15, 18 and 20 are combined in
Scheme 5.
Scheme 6. Cyclohepta[b]indole and 2-isocyanobiphenyl rearrangements.
Cyclohepta[b]indole 22 was synthesized by an im-
proved literature procedure.[25] FVP of 22 at 7508C
yielded 2% of phenanthridine 26, 45% of 2-cyanobi-
phenyl 25, and 45% of recovered starting material. When the
Next, the FVP of the authentic 2-isocyanobiphenyl 24 was
pyrolysis temperature was increased to 11008C, the conversion examined. At 8008C/10ꢀ3–10ꢀ4 hPa 2-cyanobiphenyl 25 was—
was complete, and 26 and 25 were obtained in yields of 21 as expected—formed in high yield, but, importantly, a yield of
and 35%, respectively, but, in addition, 3- and 4-cyanobiphen- about 1% of cyclohepta[b]indole 22 was also obtained
yls were formed in yields of 1.3 and 30%, respectively. These (Scheme 5). DeJong and Boyer have shown that photolysis of
results suggest that 2-cyanobiphenyl 20 was initially formed in 26 also yields 22.[25] We have shown previously that FVP and
66% yield but undergoes CN group migrations at elevated gas-phase photolysis of azulene yield almost identical results,
temperatures, and this was confirmed by re-pyrolysis of au- including the label distribution in the naphthalene formed
thentic 2-cyanobiphenyl at 11808C. High-temperature CN from 4–13C-azulene.[7]
group migrations in aromatics have been reported previous-
The reactions are interpreted in terms of the norcaradiene–
ly.[26] Acridine, benzo[f]quinoline and benzo[h]quinoline were vinylidene mechanism[7] in Schemes 5 and 6, whereby it is rec-
not detectable by gas chromatography of these pyrolysates.
ognized that the norcaradienes may be transition states rather
than intermediates.[5,6] Calculations indicate a long internal cy-
clopropane bond of about 2 ꢂ.[12,23b] It is also recog-
nised that the formation of the aromatics 2, 19 and
26 does not require the ring opening of the norcara-
dienes to vinylidenes;[6,11] but the vinylidenes (and
isocyanide in the case of 24) are required to explain
the formation of the acetylenes (and 2-cyanobiphen-
yl) as well as the reversibility of the reactions starting
from the acetylenes or isocyanide. The vinylidene–
acetylene interconversion is also necessary to explain
the formation of several labeled naphthalenes in azu-
lene rearrangements[14,15,24] and in the cyclization of
phenylbutenynes.[29]
We will now examine whether the other proposed
mechanisms of the A–N rearrangement can explain
the results described in Schemes 5 and 6. First, the
diradical mechanism of Scott and Kirms[6] is clearly
not applicable, as it would give both anthracene and
phenanthrene, and acridine and phenanthridine
(Scheme 7).
Scheme 5. Benz[a]azulene and 2-ethynylbiphenyl rearrangements
Chem. Eur. J. 2016, 22, 1 – 6
www.chemeurj.org
3
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
tial vacuum of 10ꢀ4–10ꢀ5 hPa. The pressure during pyrolysis was
usually about 10ꢀ3 hPa and the contact times of the order of ms.
The apparatus shown in Figure 3b in reference [30] was used for
preparative pyrolyses. The apparatus shown in Figure 5[30] with
a 10ꢃ0.8 cm quartz tube was used for the pyrolysis of 22 at
4008C/10ꢀ5 hPa. Products were isolated on liquid N2-cooled cold
fingers, analyzed by gas chromatography, separated by preparative
gas chromatography, and identified by direct spectroscopic com-
parison with authentic samples.
Similarly, Alder’s H-addition (spiran) and methylene-walk
pathways would give both acridine and phenanthridine
(Scheme 7). Therefore, among all the potentially viable mecha-
nisms proposed in the literature, only the norcaradiene–vinyli-
dene mechanism can explain the observed results.
In conclusion, FVP of benz[a]azulene and cyclohepta[b]in-
dole affords phenanthrene and phenanthridine, respectively, as
major products together with 2-ethynylbiphenyl and 2-cyano-
biphenyl, respectively. The latter products are seen as arising
from 1,2-rearrangements of 2-(2-biphenylyl)vinylidene and 2-
isocyanobiphenyl, respectively. The results fit nicely with the
norcaradiene–vinylidene mechanism of the azulene–naphtha-
lene rearrangement (Scheme 1).
Benz[a]azulene 15 was prepared in 0.5% yield by reaction of ethyl
diazoacetate with fluorene,[24] and in 45% yield by ring expansion
of 2-fluorenylcarbene, generated by FVP of 5-(2-fluorenyl)tetra-
zole.[22]
Cyclohepta[b]indole 22
This compound was prepared by a modification of the literature
Experimental Section
procedure.[28]
5,6,7,8,9,10-Hexahydrocyclohepta[b]indole
(5 g,
Gas chromatography was performed on a PerkinElmer 900 instru-
ment equipped with a Spectra Physics Autolab System I integrator
and SE30 or SE52 columns. Preparative gas chromatography was
carried out with a Varian Aerograph model A90P3 or an in-house
modified Gow-Mac model 69–552 instrument. Mass spectra were
obtained on a conventional sector instrument using electron ioni-
zation. IR spectra were recorded for KBr mulls or neat on KBr.
27 mmol) was dissolved in 100 mL 2-methyl-2-butanol, and 15 g
(61 mmol) of solid, powdered tetrachloro-o-benzoquinone was
added continuously with vigorous stirring over a 3 h period by
using the solid addition apparatus described for the falling solid
FVP method.[23c] After the 3 h period the resulting mixture was re-
fluxed for 2 h. After cooling to room temperature, 200 mL of dieth-
yl ether was added, and the mixture was extracted with 300 mL
2N NaOH. This extract was then extracted 12 times with 300 mL
portions of 2N HCl. Each of the green, acidic extracts were washed
The pyrolyses were performed by subliming the solid starting ma-
terials into a horizontal quartz pyrolysis tube (30ꢃ2 cm) at an ini-
Scheme 7. Potential, but non-observed, A–N rearrangement mechanisms.
&
&
Chem. Eur. J. 2016, 22, 1 – 6
www.chemeurj.org
4
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communication
[1] a) E. Heilbronner, P. A. Plattner, K. Wieland, Experientia 1947, 3, 70–71;
b) E. Heilbronner, K. Wieland, Helv. Chim. Acta 1947, 30, 947. c) E. Heil-
bronner, in Non-benzenoid Aromatic Compounds (Ed.: D. Ginsburg),
Wiley-Interscience, New York, 1959.
[2] H. Y. Cho, A. Ajaz, D. Himali, P. A. Waske, R. P. Johnson, J. Org. Chem.
2009, 74, 4137.
[3] S. Das, J. Wu, Org. Lett. 2015, 17, 5854.
[4] a) L. Brouwer, J. Troe, Internat. J. Chem. Kin. 1988, 20, 379; b) A. Laskin,
A. Lifshitz, Isr. J. Chem. 1996, 36, 257.
[5] Yu. A. Dyakov, C.-K. Ni, S. H. Lin, Y. T. Lee, A. M. Mebel, J. Phys. Chem. A
2005, 109, 8774.
[6] A. Stirling, M. Iannuzzi, A. Laio, M. Parrinello, ChemPhysChem 2004, 5,
1558.
[7] J. Becker, C. Wentrup, E. Katz, K.-P. Zeller, J. Am. Chem. Soc. 1980, 102,
5110.
[8] L. T. Scott, M. A. Kirms, J. Am. Chem. Soc. 1981, 103, 5875.
[9] This mechanism was suggested earlier for the photochemical N–A rear-
rangement: G. Kaupp, Angew. Chem. Int. Ed. Engl. 1978, 17, 150–168;
Angew. Chem. 1978, 90, 161–179.
with 50 mL of diethyl ether, then a layer of 100 mL cyclohexane
was added, and the mixture was neutralized by slow addition of
2N NaOH under vigorous stirring, which caused red coloration. The
combined cyclohexane extracts were dried over Na2SO4, filtered,
and the solution was slowly concentrated. Deep-red crystals pre-
cipitating on cooling were filtered and dried in high vacuum. Yield:
1
1.1 g (20%), m.p. 140–1418C (lit.[28] 140–1418C); H NMR (400 MHz,
CDCl3, TMS): d=8.78 (d, J(H,H)=9 Hz, 1H), 8.62 (d, J=9 Hz, 1H),
8.37 (d, J=8 Hz, 1H), 8.20 (d, J=8 Hz, 1H), 7.83–7.72 (m, 3H), 7.69
(d, J=9 Hz, 1H), 7.52 (t, J=8 Hz, 1H); 13C NMR (25 MHz, CDCl3,):
d=161.0 (s), 157.4 (s), 152.8 (d), 143.1 (s), 136.7 (d), 136.6 (d), 131.1
(d), 131.0 (d), 129.3 (d), 127.2 (s), 122.5 (d), 121.3 (d), 120.3 (d); IR
(KBr) 3040 (w), 3020 (w), 2960 (w), 2920 (w), 1615 (m), 1600 (s),
1480 (s), 1445 (s), 1400 (vs.), 1380 (m), 1325 (m), 1260 (m), 880 (s),
860 (m), 765 (vs.), 750 (s), 860 (m), 765 (vs.), 695 cmꢀ1 (vs.); MS
(70 eV): m/z (%) 180 (M+ +1, 15), 179 (11), 179 (M+, 100), 167
(100), 178 (19), 177 (9), 152 8, 151 (9), 90 (8), 89 (9).
[10] L. T. Scott, G. K. Agopian, J. Am. Chem. Soc. 1977, 99, 4506.
[11] M. J. S. Dewar, K. M. Merz, Jr., J. Am. Chem. Soc. 1986, 108, 5142.
[12] R. W. Alder, S. P. East, J. N. Harvey, M. T. Oakley, J. Am. Chem. Soc. 2003,
125, 5375.
[13] a) K.-P. Zeller, C. Wentrup, Z. Naturforsch. B 1981, 36b, 852; b) H. Gugel,
K.-P. Zeller, C. Wentrup, Chem. Ber. 1983, 116, 2775.
[14] a) R. W. Alder, G. Whittaker, J. Chem. Soc. Perkin Trans. 2 1975, 1464;
b) R. W. Alder, R. W. Whiteside, G. Whittaker, C. Wilshire, J. Am. Chem.
Soc. 1979, 101, 629.
Pyrolysis of benz[a]azulene 15
Portions of 500 mg were sublimed at 1508C and pyrolyzed at
12008C over a 3 h period to yield phenanthrene 19 (95%) and 2-
ethynylbiphenyl 18 (2%). The products were assayed by GC,
1H NMR, and mass spectrometry and identified by comparison with
authentic samples.
[15] A. Wetzel, K.-P. Zeller, Z. Naturforsch 1987, 42b, 903.
[16] L. T. Scott, Acc. Chem. Res. 1982, 15, 52.
Pyrolysis of cyclohepta[b]indole 22
[17] For a recent review of chemical activation, see: C. Wentrup, Aust. J.
Chem. 2013, 66, 852.
[18] R. Frçchtenicht, H.-G. Rubahn, J.-P. Toennies, Chem. Phys. Lett. 1989,
162, 269.
A 500 mg portion of 22 was sublimed at 140–1608C and pyrolyzed
at 7508C over a 3 h period to yield phenanthridine 26 (2%), 2-cya-
nobiphenyl 25 (45%), and recovered 22 (45%). The pyrolysis was
repeated at 11008C to yield 26 (21%) and 25 (35%), 3-cyanobi-
phenyl (1.3%) and 4-cyanobiphenyl (30%) together with a trace of
recovered starting material 22. Acridine, benzo[f]quinoline and
benzo[h]quinoline were not detectable by GC on SE52. All com-
[19] a) M. Damm, F. Deckert, H. Hippler, J. Troe, J. Phys. Chem. 1991, 95,
2005; b) M. Damm, H. Hippler, J. Troe, J. Chem. Phys. 1988, 88, 3564.
[20] a) Yu. A. Dyakov, A. M. Mebel, S. H. Lin, Y. T. Lee, C.-K. Ni, J. Chin. Chem.
Soc. 2006, 53, 161; b) M.-F. Lin, C.-L. Huang, Y. T. Lee, C.-K. Ni, J. Chem.
Phys. 2003, 119, 2032.
[21] The rearrangement also takes place upon conventional gas-phase pho-
tolysis of 4–13C-azulene with the same outcome for the labeled naph-
thalene as obtained in the FVP experiment.[7] An unlabeled A–N rear-
rangement was achieved by gas-phase infrared laser irradiation (CO2,
944 cmꢀ1) by: L. T. Scott, M. A. Kirms, B. L. Earl, J. Chem. Soc. Chem.
Commun. 1983, 1373.
[22] C. Wentrup, J. Becker, J. Am. Chem. Soc. 1984, 106, 3705.
[23] a) R. F. C. Brown, F. W. Eastwood, K. J. Harrington, G. L. McMullen, Aust. J.
Chem. 1974, 27, 2393; b) I. D. Mackie, R. P. Johnson, J. Org. Chem. 2009,
74, 499; c) C. Wentrup, J. Becker, Angew. Chem. Int. Ed. 2015, 54, 5702;
Angew. Chem. 2015, 127, 5794.
1
pounds were identified by direct comparison of GC, H NMR and
MS with authentic materials. Following pyrolysis at 4008C/
10ꢀ5 hPa, an absorption at 2120 cmꢀ1 in the IR spectrum was as-
cribed to 2-isocyanobiphenyl 24.
Pyrolysis of 2-isocyanobiphenyl 24
A 1 g portion of this compound[28] was sublimed at 2008C and py-
rolyzed at 8008C to afford 2-cyanobiphenyl 25 as the major prod-
uct together with a about 1% yield of cyclohepta[b]indole 22,
characterized by its GC retention time and color.
[24] R. W. Alder, G. Whittaker, J. Chem. Soc. Perkin Trans. 2 1975, 714.
[25] J. DeJong, J. H. Boyer, J. Org. Chem. 1972, 37, 3571.
[26] C. Wentrup, W. D. Crow, Tetrahedron 1970, 26, 4375.
[27] C. Rꢀchardt, M. Meier, K. Haaf, J. Pakusch, K. E. S. Wolber, B. Mꢀller,
Angew. Chem. Int. Ed. Engl. 1991, 30, 893; Angew. Chem. 1991, 103, 907.
[28] J. Ugi, U. Fetzer, U. Erholzer, H. Knupfer, Angew. Chem. 1965, 77, 492.
[29] a) K. Schulz, J. Hofmann, G. Zimmermann, Eur. J. Org. Chem. 1999, 3407.
b) K. Schulz, J. Hofmann, G. Zimmermann, Liebigs Ann./Recueil. 1997,
2535.
Acknowledgements
This work was supported by the Fonds der Chemischen Indus-
trie (CW 1328) and the Australian Research Council
(DP0770863).
[30] C. Wentrup, Aust. J. Chem. 2014, 67, 1150.
Keywords: arenes · flash vacuum pyrolysis · hetarenes ·
reaction mechanisms · rearrangement
Received: July 18, 2016
Published online on && &&, 0000
Chem. Eur. J. 2016, 22, 1 – 6
www.chemeurj.org
5
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Communication
COMMUNICATION
Fast and furious: Flash vacuum pyroly-
ses of the title compounds afford phen-
anthrene and phenanthridine as well as
2-ethynylbiphenyl and 2-cyanobiphenyl,
respectively. The results support the
norcaradiene–vinylidene mechanism of
the azulene–naphthalene rearrange-
ment.
&
Reaction Mechanisms
C. Wentrup,* J. Becker
&& – &&
Azulene–Naphthalene-Type
Rearrangements in Benz[a]azulene
and Cyclohepta[b]indole
&
&
Chem. Eur. J. 2016, 22, 1 – 6
www.chemeurj.org
6
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!