Purported synthesis of 3,4,7,8-tetraphenyl-1,2,5,6-tetraazocine from benzil and hydrazine: Competing cyclization and carbon-carbon σ-bond scission

Article abstract of DOI:10.1002/ejoc.200700848

The claims that 3,4,7,8-tetraphenyl-1,2,5,6-tetraazocine can be prepared by the thermal condensation of two moles of benzil monohydrazone or of an equimolar mixture of benzil and benzil dihydrazone have been thoroughly reinvestigated. When such thermolyses were conducted in moist air, neither the claimed 3,4,7,8-tetraphenyl-1,2,5,6-tetraazocine nor the precedented isomeric tetraazapentalene derivative was detected. The following products were unambiguously formed from the heating of molten benzil monohydrazone (%): benzil (10), benzaldehyde (10), benzamide (22), benzyl phenyl ketone (19), benzil bis(ketazine) (11), 3,4,5,6-tetraphenylpyridazine (9), benzil benzaldehyde azine (10), and, after column chromatography, 2,4,5-triphenylimidazole (2). This last component had a melting point and the fluorescent properties in UV light attributed by the original investigator to the mistakenly presumed 3,4,7,8-tetraphenyl-1,2,5,6-tetraazocine. Thus, the original claims for the synthesis of such a novel tetraazocine ring or even for the synthesis of the precedented isomeric zwitterionic tetraazapentalene have now been repudiated. The formation of 2,4,5-triphenylimidazole as a side reaction in the thermolysis of benzil monohydrazone can readily be rationalized as arising from benzil, benzaldehyde, and a source of ammonia, namely, benzamide, in the long-known Radziszewski reaction. Corroborating evidence was provided by data from the thermolysis of benzil dihydrazone. In addition, the origin of other side products is explained in terms of other possible condensations. Finally, preliminary experiments on using irreversible dehydrating agents such as titanium(IV) isopropoxide with benzil monohydrazone indicate that 3,4,7,8-tetraphenyl-1,2,5,6-tetraazocine is formed at room temperature as a transitory intermediate, which eliminates dinitrogen to produce 3,4,5,6-tetraphenylpyridazine. Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

Full text of DOI:10.1002/ejoc.200700848

FULL PAPER
DOI: 10.1002/ejoc.200700848
Purported Synthesis of 3,4,7,8-Tetraphenyl-1,2,5,6-tetraazocine from Benzil
and Hydrazine: Competing Cyclization and Carbon–Carbon σ-Bond Scission[‡]
John J. Eisch,*[a] Tsz Y. Chan,[a] and John N. Gitua[b]
Keywords: Dimerization / Cyclization / Nitrogen heterocycles / C–C bond scission
The claims that 3,4,7,8-tetraphenyl-1,2,5,6-tetraazocine can
be prepared by the thermal condensation of two moles of
benzil monohydrazone or of an equimolar mixture of benzil
and benzil dihydrazone have been thoroughly rein-
vestigated. When such thermolyses were conducted in moist
air, neither the claimed 3,4,7,8-tetraphenyl-1,2,5,6-tetraazoc-
ine nor the precedented isomeric tetraazapentalene deriva-
tive was detected. The following products were unambigu-
ously formed from the heating of molten benzil monohydraz-
one (%): benzil (10), benzaldehyde (10), benzamide (22),
benzyl phenyl ketone (19), benzil bis(ketazine) (11), 3,4,5,6-
tetraphenylpyridazine (9), benzil benzaldehyde azine (10),
and, after column chromatography, 2,4,5-triphenylimidazole
(2). This last component had a melting point and the fluores-
cent properties in UV light attributed by the original investi-
gator to the mistakenly presumed 3,4,7,8-tetraphenyl-
1,2,5,6-tetraazocine. Thus, the original claims for the synthe-
sis of such a novel tetraazocine ring or even for the synthesis
of the precedented isomeric zwitterionic tetraazapentalene
have now been repudiated. The formation of 2,4,5-tri-
phenylimidazole as a side reaction in the thermolysis of ben-
zil monohydrazone can readily be rationalized as arising
from benzil, benzaldehyde, and
a source of ammonia,
namely, benzamide, in the long-known Radziszewski reac-
tion. Corroborating evidence was provided by data from the
thermolysis of benzil dihydrazone. In addition, the origin of
other side products is explained in terms of other possible
condensations. Finally, preliminary experiments on using
irreversible dehydrating agents such as titanium(IV) isoprop-
oxide with benzil monohydrazone indicate that 3,4,7,8-tet-
raphenyl-1,2,5,6-tetraazocine is formed at room temperature
as a transitory intermediate, which eliminates dinitrogen to
produce 3,4,5,6-tetraphenylpyridazine.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2008)
Introduction
Our long-standing interest in open-chain and cyclic con-
jugated imines[1,2] has drawn our attention to two related
brief communications describing the thermal autoconden-
sation of either benzil monohydrazone (1)[3] or of a 1:1 mix-
ture of benzil (2a) and benzil dihydrazone (2b)[4] to the pur-
ported 3,4,7,8-tetraphenyl-1,2,5,6-tetraazocine (3), but in
very low yield and with inadequate experimental detail to
support their structural assignment[5] (Scheme 1).
Further work with the α-hydrazone of either the methyl
ester of benzoylglyoxalic acid (4a) or of the ethyl ester of
α,β-diketobutyric acid (4b) did provide reliable analytical
data that such an autocondensation might have led to the
structure of the corresponding tetraazocine (5a or 5b)
[Equation (1)].[6,7] Such a finding tended to enhance the
credibility of the conclusions drawn from the earlier work
with benzil hydrazones 1 and 2b.
Scheme 1.
[‡] Studies in Non-Pyridinoid Aza-Aromatic Systems, 7. Part 6:
Ref.[1]
(1)
[a] The State University of New York at Binghamton, Department
of Chemistry,
Structural assignments in these systems became more
complicated after an X-ray crystallographic investigation of
the diacid derived from the saponification of 5a and subse-
quent acidification.[8] This diacid was found to possess not
P. O. Box 6000, Binghamton, New York 13902-6000, USA
Fax: +1-607-777-4865
E-mail: jjeisch@binghamton.edu
[b] Drake University, Chemistry Department,
207 University Avenue, Des Moines, Iowa 50311, USA
392
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 392–397

3,4,7,8-Tetraphenyl-1,2,5,6-tetraazocine from Benzil and Hydrazine
a tetraazocine nucleus (6a) but the structure of a novel aro-
matic zwitterionic 1,3a,4,6a-tetraazapentalene (6b). If tetra-
azocine 5a were actually an intermediate, it must have iso-
merized during the reaction. Although the structures of di-
acid 6b and its ester 5a are then no longer in doubt, the
crystallographic group nevertheless opined that the product
from benzil monohydrazone probably is that of tetraazocine
3.[8]
(2)
What aroused our scepticism concerning the purported
tetraazocine 3 was the absence of verifiable analytical data,
the lack of any color, and the high melting point (276 °C).
All aryl azomethine or hydrazones we have encountered are
distinctly yellow and melt under 200 °C. Therefore, we have
launched a reinvestigation of the thermal reactions of hy-
drazones 1 and 2. This study has led to the surprising con-
clusions that cyclization and carbon–carbon σ-bond scis-
sions of such hydrazones begin as low as 50 °C and that
purported 3 has neither the structure of a tetraazocine nor
Scheme 2.
Thermal Decomposition of Benzil Monohydrazone at
180 °C in Moist Air (Presumed Metze Conditions)
of
a tetraazapentalene, but is the known 2,4,5-tri-
Benzil monohydrazone (1) was heated in open air at
180 °C for 8 h. The reaction product, 95% of the original
weight, consisted of 19% of 7, 11% of 8, 9% of 3,4,5,6-
tetraphenylpyridazine (13), 10% of benzaldehyde benzil az-
ine (14), 19% of benzaldehyde (15), 10% of benzil (2a), and,
remarkably, 22% of benzamide (16) [Equation (3)].
phenylimidazole. Therewith, the joint claim of Metze and
of Schlesinger to the synthesis of a tetraazocine has been
decisively repudiated.
Results and Discussion
Thermal Stability of Benzil Monohydrazone (1)
Pertinent to the thermal decomposition of 1 are two ob-
servations: (1) heating a neat sample of 1 for 2 h under re-
duced pressure (14 Torr) caused partial decomposition to
benzyl phenyl ketone, which distilled off, and known benzil
bis(ketazine) (8), which remained as a residue [Equa-
tion (2)];[9] and (2) in this study, heating 1 in 95% ethanol
at 110° for 17 h led to the 75% decomposition of 1 and
products that consisted of 24% of 7 and 25% of 8 [cf. Equa-
tion (2)], as well as 7% of benzil (2a), 15% of ethyl benzoate
(9), and 4% of trans,trans-benzaldehyde azine (12). The ori-
(3)
Most noteworthy are products 14, 15, and 16, which
gin of 7 in Equation (2) can be attributed to a Wolff– must have arisen from C–C bond scission of 1. Moreover,
Kishner reaction but the presence of ethyl benzoate (9) and the formation of 16 proves that N–N bond homolysis must
azine 12 reveals the occurrence of a C–C bond solvolysis of also have occurred. There is a straightforward explanation
1, possibly via 10, to produce 9 and benzaldehyde hydraz- for the formation of 14: when a 1:1 mixture of 1 and 15 in
one (11); the latter compound can disproportionate to azine 95% ethanol was stored at 0 °C, a 70% yield of benzalde-
12 and hydrazine (Scheme 2). The source of ketazine 8 will hyde benzil azine (14) was deposited [Equation (4)]. The
be considered later. These reactions show that intermo- origin of 13, however, is puzzling and its manner of forma-
lecular reactions of 1 can occur at lower temperatures and tion receives due consideration when the possible role of
that C–C bond cleavage occurs readily even in warm tetrahydrotetraazocine derivatives in these reactions is dis-
alcohol.
cussed (cf. infra).
Eur. J. Org. Chem. 2008, 392–397
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
393

J. J. Eisch, T. Y. Chan, J. N. Gitua
FULL PAPER
Recrystallization from ethanol gave a melting point of 275–
276 °C.
The original ethanolic filtrate was evaporated and ana-
1
lyzed by H and 13C NMR spectroscopy. The presence of
62% of benzyl phenyl ketone (7), 5% of trans,trans-benzal-
dehyde azine, and 33% of 2b was established.
(4)
When an identical thermolysis of 2b was carried out, but
in the presence of a threefold excess of benzaldehyde, about
a 3.5% yield of 17 was achieved.
Finally, upon column chromatography of the entire reac-
tion mixture and the elution of all the foregoing compo-
nents, a colorless, slowest-moving component (17) on the
column or on TLC was detected by its pronounced bright-
blue fluorescence. Upon isolation, 0.20 mg of compound 17
had a melting point of 274–276 °C (from ethanol). Al-
though the properties of 17 resemble those reported for 3,
discussion of its identity will be deferred at this point until
other data are presented.
Identity and Origin of Compound 17
The melting point, migration on TLC plates, and infra-
1
red, H and 13C NMR spectra of compound 17 isolated
from the foregoing experiments was compared with an au-
thentic sample of 2,4,5-triphenylimidazole (m.p. 275–
276 °C), and it was found to be identical in all respects. The
melting point of a mixture of 17 and the authentic imid-
azole was not depressed.
Thermal Decomposition of Benzil Monohydrazone at
180 °C Under a Dry Argon Atmosphere
The origin of 2,4,5-triphenylimidazole (17) in the pyroly-
sis of benzil monohydrazone (1) or of benzil dihydrazone
(2b) conducted in moist air is not difficult to trace. As
shown in Equation (3), monohydrazone 1 produces benzil
(2a), benzaldehyde (15), and benzamide (16), a source of
ammonia. More than 125 years ago, Radziszewski demon-
strated that passing ammonia into an ethanolic solution of
2a and 15 at 50 °C leads to a quantitative formation of
2,4,5-triphenylimidazole (17) [Equation (6)].[10]
The decomposition of hydrazone 1, carried out under
identical conditions except that the atmosphere was anhy-
drous argon, gave the following array of products: 16% of
7, 11% of 8, 9% of 13, 31% of 14, and 23% of a new
product, benzil (benzyl phenyl ketone) azine (18). The for-
mation of 18 can readily be ascribed to the thermal conden-
sation of 1 and 7 [Equation (5)].
(5)
(6)
Further noteworthy are the following aspects: (1) the rel-
ative amounts of 7, 8, and 13 formed in the two modes of
decomposition, in moist air and under an argon atmo-
sphere, are comparable: 7: 16 vs. 19%, 8: 18 vs. 11%, and
13: 12 vs. 9%; (2) under an argon atmosphere the formation
of 18 is accompanied by a great increase in the proportion
of 14, from 10 to 23%; and (3) under an argon atmosphere
products such as benzaldehyde (15), benzamide (16), and
17 are present only in trace amounts. Because potential
products such as 15, 16, and 17 stem from C–C, C–N, or
N–N bond cleavages of 1 during decomposition, it can be
concluded that the moist air present under Metze’s condi-
tions was largely responsible for such C–C and N–N bond
cleavages of 1.
Origins of Benzil Bis(ketazine) (8) and 3,4,5,6-
Tetraphenylpyridazine: Transitory Formation of
Tetrahydrotetraazocine Intermediates
As mentioned above (Scheme 2), ketazine 8 forms readily
from benzil monohydrazone (1) even in warm ethanol.
Most likely 1 undergoes cyclic dimerization to the tetra-
hydrotetraazocine derivative 19, which by consecutive loss
and readdition of H2O would isomerizes to conjugated iso-
mer 20. Structure 20 could readily eliminate hydrazine to
yield 8 (Scheme 3). The complete loss of H2O from 19 or 20
to produce 3 would undoubtedly involve a high activation
energy, because as molecular models show, 3 would assume
a nonplanar, highly strained tub-shaped conformation, like
cyclooctatetraene itself. We conclude, therefore, that the
Thermal Decomposition of Benzil Dihydrazone at 180 °C in
Moist Air
The decomposition of dihydrazone 2b at 180 °C in moist prospects of synthesizing tetraazocine 3 by the ordinary de-
air was cleaner and more informative. The reaction product hydration of 1 are unpromising, as the water eliminated
dissolved in 95% ethanol deposited very small amounts of would be available for the reversible rehydration of interme-
compound 17 with
a melting point of 274–276 °C. diates, such as in steps 19 Ǟ 20 (Scheme 3).
394
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 392–397

3,4,7,8-Tetraphenyl-1,2,5,6-tetraazocine from Benzil and Hydrazine
Scheme 3.
Scheme 5.
To circumvent such reversible dehydration–rehydration
in the dimerization of benzil monohydrazone (1), we have
treated 1 at room temperature in toluene solution with
0.5 equiv. of titanium(IV) isopropoxide (21) and catalytic
amounts of p-toluenesulfonic acid. Subsequent hydrolysis
showed that 59% of 1 had been consumed and that 31% of
7 and 28% of 13 had been formed. One straightforward
interpretation of this outcome would be that intermediate
19 would have undergone irreversible dehydration with 21
to form 3. Because of the ring strain and antiaromaticity of
such a planar ring, 3 would rapidly undergo ring closure to
fused ring 22, just as cyclooctatetraene has been shown to
do. However, 22 then would undergo facile loss of N2 with
the generation of azaaromatic 13 (Scheme 4).[11]
(7)
Conclusions
A thorough reinvestigation of the claims that the thermo-
lyses of benzil monohydrazone (1) and of benzil dihydra-
zone (2a) in moist air lead to 3,4,7,8-tetraphenyl-1,2,5,6-
tetraazocine (3) has shown that neither this tetraazocine nor
the isomeric tetraphenyl-1,3a,4,6a-tetraazapentalene is
formed in these reactions. In both thermolyses, the pur-
ported 3 has been unambiguously identified as 2,4,5-tri-
phenylimidazole (17). The origin of 17 was established by
identifying C–C and N–N bond cleavage side products,
such as, benzil, benzaldehyde, and benzamide, a source of
ammonia. In the classic Radziszewski reaction such compo-
nents readily combine to form 17.
Preliminary results support the hope that thermally cy-
clic imines may be procured by irreversible dehydrations of
carbonyl adducts of amines or hydrazines with Ti(OiPr)4 at
lower temperatures, where such undesired cleavages occur
less readily.
Scheme 4.
Experimental Section
In an analogous manner, Wolff–Kishner product 7 could
arise from the cyclizing action of Ti(OiPr)4 and acid cata-
lyst on monomer 1 to form titanate ester 23, which would
readily lose N2 to yield 24. Hydrolysis would produce ben-
zyl phenyl ketone (7) (Scheme 5).
A plausible alternative route to 13 might be that 1 and 7
would react according to Equation (5) to form 18, which
could then undergo intramolecular dehydration to produce
13 [Equation (7)]. But opposing this alternative path to 13
are two observations: (1) none of proposed intermediate 18
was detectable from the reaction of 1 with Ti(OiPr)4 at
room temperature; and (2) prolonged heating of 18 at
180 °C leads only slowly to 13. Further research will explore
the implications of our findings for the synthesis of labile
cyclic imines.
Instrumentation and Analysis: Some reactions were carried out un-
der a positive pressure of anhydrous, oxygen-free argon.[12] Other
reactions were carried out in a flask equipped with an air- or water-
cooled condenser open to the surrounding moisture (“Metze condi-
tions”). The IR spectra were recorded with a Perkin–Elmer instru-
ment (model 457) and samples were measured either as mineral oil
mulls or as KBr films. The NMR spectra (1H and 13C) were re-
corded with a Bruker spectrometer (model EM-360) and tetrameth-
ylsilane (Me4Si) was used as the internal standard. The chemical
shifts reported are expressed on the δ scale in parts per million
(ppm) from the Me4Si reference signal. All 1H and 13C NMR spec-
tra were recorded at 360 MHz and in CDCl3. The J values are
not given when they fall in the customary 2–7 Hz range, and the
multiplicity of a signal is only indicated when there is baseline sepa-
ration from other signals; otherwise, the multiplicity is given as
Eur. J. Org. Chem. 2008, 392–397
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
395

J. J. Eisch, T. Y. Chan, J. N. Gitua
FULL PAPER
“m”. The GC–MS measurements and analyses were performed
could be isolated from the aqueous extract; the acid could arise by
with a Hewlett–Packard GC 5890/Hewlett–Packard 5970 mass-se- oxidation of 15 or hydrolysis of 16 during the pyrolysis.
lective-detector instrument. The gas chromatographic analyses were
3,4,5,6-Tetraphenylpyridazine (13): M.p. 192–193 °C. 1H NMR: δ =
performed with a Hewlett–Packard instrument (model 5880) pro-
7.41–7.31 (m), 7.29–7.20 (m), 7.12–7.03 (m), 6.87–6.84 (d, 2 H)
vided with a 2-m Ov-101 packed column or with a Hewlett–Pack-
ppm. 13C NMR: δ = 158.9, 138.1, 137.4, 135.2, 130.4, 130.1, 128.2,
ard instrument (model 4890) having a 30-m SE-330 capillary col-
127.81, 127.80, 127.4 ppm.
umn. Melting points were determined with a Thomas-Hoover
Unimelt capillary melting point apparatus and are uncorrected.
Benzaldehyde Benzil Azine (14): M.p. 139–141 °C. 1H NMR: δ =
8.58 (2 H), 7.97–7.95 (d, 2 H), 7.86–7.84 (d, 2 H), 7.60–7.25 (m)
Starting Reagents and Reaction Products: The following com-
pounds employed in this study, benzil monohydrazone (1), benzil
dihydrazone (2b), benzyl phenyl ketone (7), benzil (2a), ethyl benzo-
ppm. 13C NMR: δ = 197.5, 1617.4, 162.1, 135.4, 133.8, 133.6,
132.6, 131.5, 131.4, 129.2, 128.9, 128.87, 128.85, 128.5, 127.8 ppm.
ate (9), trans,trans-benzaldehyde azine (12), benzaldehyde (15),
benzamide (16), and 2,4,5-triphenylimidazole (17) were commer-
cially available in at least 98% purity. Benzil bis(ketazine) (8), a
previously characterized compound, was prepared according to a
literature reference.[9] The previously reported 3,4,5,6-tetraphen-
ylpyridazine (13) was synthesized by heating an equimolar mixture
of 1 and 7 at 180 °C for 8 h. Column chromatography on silica gel
yielded 13% of 13 (m.p. 191–193 °C) with MS (electrospray with
Subsequent column chromatography of the product on silica gel
with hexane as the eluent permitted the isolation of 0.20 mg of
colorless solid 17, which proved to be 2,4,5-triphenylimidazole
(m.p. 275–276 °C) by spectral and mixture m.p. criteria.
Thermal Decomposition of Benzil Monohydrazone (1) Under a Dry
Argon Atmosphere: When the foregoing heating of a 1.0-gram sam-
ple of 1 was repeated under a dry argon atmosphere, some products
clearly due to C–C or N–N bond cleavage, namely, 15, 16, and 17,
were sharply reduced in amount or not at all detected. One product
stemming from the C–C bond cleavage in 1 was increased in relative
amount: benzaldehyde benzil azine (14) rose from 10 to 31%.
NaI) 384.1616 (with Na+); calcd. 384.1614; and correct 1H and 13
C
NMR spectra. The other product, 87%, was the cross-ketazine
from 1 and benzyl phenyl hydrazone. Finally, the cross-ketazine
from 1 and 11, namely, 14, was prepared in 5 mL of 95% ethanol
by mixing 250 mg (1.11 mmol) of 1 with 3.0 mL (an excess) of 15.
Storing the solution at 0 °C deposited 340 mg (70%) of yellow 14,
m.p. 139–141 °C.
Analysis: benzaldehyde benzil azine (14), 31%; benzil (benzyl
phenyl ketone) azine (18), 23%; benzyl phenyl ketone (7), 16%;
3,4,5,6-tetraphenylpyridazine (13), 12%; and benzil bis(ketazine)
(8), 18%.
Analysis and Identification of Reaction Products: From the forego-
ing known samples, their respective melting points, their 1H and
13C NMR spectra, and their behavior upon thin-layer chromatog-
raphy were available criteria for the identification and quantifica-
tion of the various reaction products.
Benzil (benzyl phenyl ketone) Azine (18): M.p. 106–108 °C. 1H
NMR: δ = 7.91–7.82 (m), 7.55–7.12 (m), 4.54 (s, 2 H) ppm. 13C
NMR: δ = 198.0, 165.9, 164.8, 137.2, 136.5, 135.4, 133.8, 132.9,
131.2, 130.2, 129.0, 128.8, 128.7, 128.6, 128.1, 127.7, 127.6, 126.2,
33.8 ppm.
1
The H and 13C NMR spectra of most of the known products are
conveniently available from compilations of the Aldrich Chemical
Company.[13] Where such spectra are not readily procurable, the
relevant 1H and 13C NMR spectroscopic data are given here in the
individual experimental procedures.
Thermal Decomposition of Benzil Dihydrazone (2b) at 200 °C in
Moist Air: Benzil dihydrazone (2b, 1.00 g, 4.20 mmol) was heated
in open air in a flask at 200Ϯ10 °C. A yellow solid (0.81 g) re-
mained after loss of volatiles. About 30% of 2b remained with the
formation of 60% of benzyl phenyl ketone (7) and 10% of
trans,trans-benzaldehyde azine (12). TLC of the entire reaction
product showed the presence of small amounts of 2,4,5-tri-
phenylimidazole (17) by its retention time and its bright blue fluo-
rescence with UV light.
Experimental Procedures for the Reactions (seriatim as given in the
Results)
Benzil Monohydrazone (1) in Refluxing Ethanol: A solution of 1
(1.00 g, 4.46 mmol) in 95% ethanol (20 mL) was heated at reflux
(bath: 110 °C) for 17 h. Dilution with water and extraction with
ether gave an organic layer, which was separated and dried with
anhydrous Na2SO4. Solvent evaporation gave 0.70 g of residue that
When an identical thermolysis of 2b was done, but in the presence
of benzaldehyde (12.6 mmol), similar proportions of 2b, 7, and 12
were found. But after extraction of the product with CHCl3, 35 mg
of a white solid remained, m.p. 274–276 °C, which proved to be
1
was directly analyzed by H and 13C NMR spectroscopy to yield
the composition of 24% of benzyl phenyl ketone (7), 25% of benzil
bis(ketazine) (8), 7% of benzil (2a), 15% of ethyl benzoate (9), and
4% of trans, trans-benzaldehyde azine (12).
1
2,4,5-triphenylimidazole (mixture m.p. and H and 13C NMR and
IR spectral comparisons).
Dehydration of Benzil Monohydrazone (1) with Titanium(IV) Isoprop-
oxide at Room Temperature: A solution of dried benzil mono-
hydrazone (1, 2.00 g, 8.92 mmol), titanium(IV) isopropoxide
(1.33 mL, 4.46 mol), and p-toluenesulfonic acid (0.10 g, 0.5 mmol)
in dried toluene (100 mL) was stirred for 24 h at room temperature
Benzil Bis(ketazine) (8): M.p. 202–204 °C. 1H NMR: δ = 7.97–7.94
(d, 2 H), 7.61–7.58 (d, 2 H), 7.53–7.48 (m), 7.38–7.34 (m), 7.27–
7.23 (m) ppm. 13C NMR: δ = 197.4, 167.0, 135.5, 134.1, 132.2,
131.7, 129.2, 129.0, 128.7, 128.1 ppm.
Thermal Decomposition of Benzil Monohydrazone (1) at 180 °C in during which time the mixture turned a bright orange color. Ad-
Moist Air: A dry sample of 1 (1.00 g, 4.46 mmol) in an open flask
was heated at 180 °C (bath) for 8 h. The brown residue (0.95 g) was
dissolved directly in CDCl3 and analyzed by 1H and 13C NMR
spectroscopy. Spectral analysis led to a composition of 10% of ben-
zil (2a), 19% of benzyl phenyl ketone (7), 8% of benzil bis(ketazine)
(8), 9% of 3,4,5,6-tetraphenylpyridazine (13), 10% of benzaldehyde
benzil azine (14), 22% of benzamide (16), and 19% of benzalde-
hyde (15). It should be noted that minor amounts of benzoic acid
dition of H2O (15 mL) caused the color to change to dark brown
and a solid to form. Filtration of the suspension, extraction of the
filtrate with ethyl ether (3ϫ15 mL), drying the ether extracts with
anhydrous Na2SO4, and removal of the volatiles gave 1.6 g of
brown residue. TLC and 1H and 13C NMR spectral analyses
showed that the residue consisted of 31% of benzyl phenyl ketone
(7), 28% of 3,4,5,6-tetraphenylpyridazine (13) and 41% of starting
material 1.
396
www.eurjoc.org
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2008, 392–397

3,4,7,8-Tetraphenyl-1,2,5,6-tetraazocine from Benzil and Hydrazine
[5] Metze makes the following statements in ref.[3] without giving
any experimentally determined values or analyses: (1) “Analyse
und Molekulargewichtsbestimmung ergaben die Bruttofomular
C28H20N4”; and (2) “Die Substanz..wird erst durch halbkon-
zentrierte Schwefelsäure bei 210–220 °C unter Druck hy-
drolytisch in Benzil und Hydrazin aufgespalten im Mol-
verhältnis 1:1”. Both these statements would require actual ex-
perimental values and detail to be considered scientifically
credible.
Acknowledgments
This research was conducted with the financial support of a Senior
Scientist Award to J. J. E. from the Alexander von Humboldt Stif-
tung of Bonn, Germany. Valuable discussions with and commen-
tary from Professor Udo Brinker of the University of Vienna, Aus-
tria, are gratefully noted.
[6] R. Pfleger, H.-G. Hahn, Chem. Ber. 1957, 90, 2411–2416.
[7] R. Pfleger, E. Garthe, K. Rauer, Chem. Ber. 1963, 96, 1827.
[8] M. Brufani, G. Casini, W. Fedeli, F. Mazza, A. Vaciago, Gazz.
Chim. Ital. 1971, 101, 322–343.
[1] J. J. Eisch, T. Abraham, Tetrahedron Lett. 1976, 17, 1647–1650.
[2] Novel conjugated imines are of ongoing importance in our or-
ganometallic chemistry both as ligands for subvalent transi-
tion-metal complexes (cf. inter alia, J. J. Eisch, X. Ma, K. I.
Han, J. N. Gitua, C. Krueger, Eur. J. Inorg. Chem. 2001, 77–
88) and as substrates for epimetalation or reductive dimeriza-
tion (cf., e.g., J. J. Eisch, J. R. Alila, Organometallics 2000, 19,
1211–1213).
[3] R. Metze, Angew. Chem. 1956, 68, 580–581. Compound 3 was
claimed to have been isolated, in low yields (≈ 10%), from: (1)
reaction of 1 in formamide at 150 °C as a side product; (2)
heating 1 above its m.p.; and (3) heating 1 in ethanol at 170–
180 °C in a sealed tube. Compound 3 is said to be a colorless
product that crystallizes from benzene as needles melting at
278 °C, which fluoresce a bright blue under UV light.
[4] H. Schlesinger, Angew. Chem. 1960, 72, 563. Compound 3 was
said to have resulted in 12% yield from heating a 1:1 mixture
of 2a and 2b at 130 °C for 3 h or by heating the identical 1:1
mixture at reflux in ethylene glycol monomethyl ether for 45 h.
From benzene, 3 was crystallized (m.p. 285–286 °C). A sample,
said to have been prepared according to Metze’s method,
melted also at 285–286 °C and the mixture m.p. of the two
samples was undepressed. Schlesinger accepted Metze’s struc-
ture for 3 without question and without additional evidence.
Schlesinger also let substituted derivatives of 2a and 2b react
and again simply assumed that the products were the corre-
sponding tetraazocines.
[9] H. Staudinger, O. Kupfer, Ber. Dtsch. Chem. Ges. 1911, 44, 22.
[10] B. Radziszewski, Ber. Dtsch. Chem. Ges. 1882, 15, 1493–1496.
[11] For initial evidence leading to a postulated equilibrium be-
tween cyclooctatetraene and its bicyclo[4.2.0]octatriene isomer
and a similar equilibrium between 1,3,5-cyclooctatriene and its
bicyclo[4.2.0]octadiene isomer, refer to: E. Müller, Neuere An-
schauungen der Organischen Chemie, 2nd ed., Springer, Berlin,
1957, pp. 354–356, and the primary references to the works of
Reppe, Cope, and Alder cited therein. It is interesting to note
that such reports of the strain and nonplanarity of the eight-
membered C8-ring appeared before Metze and Schlesinger
published their postulated “planar” tetraazocine ring without
misgivings.
[12] Detailed descriptions of useful techniques and procedures for
maintaining and working under an anhydrous and inert atmo-
sphere are given in: J. J. Eisch, Organometallic Syntheses, Aca-
demic, New York, 1981, vol. 2, pp. 3–52.
1
[13] J. C. Pourhert, J. Behnke (Eds.), Aldrich Library of 13C and H
FT-NMR Spectra, Aldrich Chemical Company, Milwaukee,
Wisconsin, USA, 1993, vols. 1,2.
Received: September 10, 2007
Published Online: November 2, 2007
Eur. J. Org. Chem. 2008, 392–397
© 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
397