Oxidative benzylic substitution of azulene-1-azo(4′-methylbenzene)s

Article abstract of DOI:10.1039/a908862h

Treatment of azulene-1-azo(4′-methylbenzene) with position 3 blocked by different substituents X (2) or by steric hindrance (3), with FeCl₃ in benzene, resulted in the substitution of one or two benzyl hydrogens by phenyl groups. Starting either from 2 (X = CH₃, Cl) or from 3, triarylmethane derivatives 7 or 10 and corresponding alcohols 9 or 11 are the main separated products while starting from 2 (X = OCH₃), diarylmethanol 8 as well as the related benzophenone and ether, 13 and 14, respectively, are isolated as major products. Diphenylmethane and triphenylmethane also result from almost all reactions. Conversion of the starting materials was good and the products that were separated and characterised are obtained in 43.5-60.5% overall yields. Single-electron transfer oxidation was thought to be the first reaction step with the generation of the radical cation 2^·+. As a result of the proton and electron transfer or hydrogen atom transfer from 2^·+, the cation 16 with some benzylic character was formed. The cation 16 alkylated benzene in the subsequent reaction and diarylmethane derivatives were formed. A similar reaction pathway can be proposed for triarylmethane products generation.

Full text of DOI:10.1039/a908862h

View Article Online / Journal Homepage / Table of Contents for this issue
Oxidative benzylic substitution of azulene-1-azo(4Ј-methyl-
benzene)s
Alexandru C. Razus* and Carmen Nitu
1
Institute of Organic Chemistry “C. D. Nenitzescu” of Romanian Academy,
Spl. Independentei 202 B, PO Box 15-258, 71141-Bucharest, Romania. Fax: ϩ40 1 3121601;
E-mail acrazus@ccoux.cco.ro
Received (in Cambridge, UK) 8th November 1999, Accepted 19th January 2000
Treatment of azulene-1-azo(4Ј-methylbenzene) with position 3 blocked by different substituents X (2) or by
steric hindrance (3), with FeCl3 in benzene, resulted in the substitution of one or two benzyl hydrogens by phenyl
groups. Starting either from 2 (X = CH3, Cl) or from 3, triarylmethane derivatives 7 or 10 and corresponding
alcohols 9 or 11 are the main separated products while starting from 2 (X = OCH3), diarylmethanol 8 as well as
the related benzophenone and ether, 13 and 14, respectively, are isolated as major products. Diphenylmethane
and triphenylmethane also result from almost all reactions. Conversion of the starting materials was good and the
products that were separated and characterised are obtained in 43.5–60.5% overall yields. Single-electron transfer
oxidation was thought to be the first reaction step with the generation of the radical cation 2 ϩ. As a result of the
᝽
proton and electron transfer or hydrogen atom transfer from 2 ϩ, the cation 16 with some benzylic character was
᝽
formed. The cation 16 alkylated benzene in the subsequent reaction and diarylmethane derivatives were formed.
A similar reaction pathway can be proposed for triarylmethane products generation.
In a recent study on the azulene-1-azobenzenes 1 oxidation
in the presence of ferric chloride in benzene,1 we have found
that the reaction was mainly directed to 4Ј,4Ј or 3,3 coupling
depending on the structure of 1. Whereas for 1 R = H, the
our studies on single-electron oxidation of the compounds 1
R = CH3, in the absence of a classical nucleophile, the reagents
which can attack the benzyl position were azulene for 1 X = H
and benzene. Based on these results we think we could obtain
interesting information on the benzyl substitution from the
behaviour of compounds 1 R = CH3 with the reactive position 3
blocked in order to avoid undesirable reactions.
Results and discussion
The starting substrates were 3-substituted azulene-1-azo(4Ј-
methylbenzene)s, 2a–2c; the effect of steric hindrance of the 3
position as in compound 3 was also considered.
oxidative p-phenyl coupling was the major reaction pathway,
the dimerisation of 1 R ≠ H occurred mainly on position 3 of
the azulene moiety. The compound 1 R = OCH3 reacted nearly
quantitatively and the separated and identified coupling
products exceeded 70% yield. Surprisingly, in spite of the good
reaction conversion of 1 R = CH3, the reaction mixture con-
tained 3,3-coupling product only in 14% yield, together with a
1
significant amount of tar. In the complicated H-NMR spec-
trum of a mixture extracted from this tar, the singlet signal for
4Ј-methyl protons (at 2.40 ppm) disappears and some small
singlets can be observed between 4 and 6 ppm. It seems, there-
fore, that the azulene-1-azo(4Ј-methylbenzene) reacted not only
at position 3 in the azulene moiety but also at the benzyl methyl
to promote the oligomerisation and polymerisation.
For the synthesis of these compounds, the corresponding
substituted azulenes were coupled with the diazonium salt of
toluidine (in 90–97% yields).6
The oxidation of the azo compounds was carried out in
benzene, using ferric chloride in excess (from 4:1 to 6:1 with
respect to the azo compound), at room temperature, with stir-
ring. The reaction proceeded slowly and by prolonging the
reaction time, modest yields were accompanied by many
secondary reactions. After work-up, the products were separ-
ated by column chromatography and, when necessary, purified
by crystallisation (some chromatographic fractions were con-
taminated by traces of compounds with aliphatic protons in the
1H-NMR spectrum). The main reaction products are reported
in Scheme 1 and in Table 1.
The substitution of the benzyl position under oxidative con-
ditions with a nucleophile is one of the important reactions of
alkylaromatics from both the practical and the theoretical point
of view, and radical cations are proved to be the reaction inter-
mediates.2 In most cases, polyalkylated benzenes or anthracenes
were substrates and the pyridine bases or other species with N
or O functions present in medium were tested as nucleophiles.
4
In the presence of Cu()/S2O8Ϫ,3 Fe()/H2O2 or tris(phenan-
throline)iron tris(hexafluorophosphate),5 benzyl alcohols and
carbonyl compounds were reported as reaction products. In
DOI: 10.1039/a908862h
J. Chem. Soc., Perkin Trans. 1, 2000, 989–994
This journal is © The Royal Society of Chemistry 2000
989

View Article Online
Table 1 Oxidation of azulene-1-azo(4Ј-methylbenzenes) 2 and 3 with FeCl3 in benzenea
Product yieldb,c (%)
Recovered
X
2 or 3 (%)
4
5
6
7
8
9
10
11
CH3 (2a)
Cld (2b)
OCH3 d (2c)
3
35
40
14.5
23.5
2
4.5
—
3
3
13
0
11
2
0
16
0
0
30
20
22
—
—
—
21
—
—
—
17
e
e
0
—
0f
11
—
—
—
a The contact time between the reactants was 48 hours for 2a, 2b and 3 and 4 hours for 2c. b Yields in %, based on recovered starting material.
1
c For each experiment, the final eluted fraction (with very polar solvents) shows in the aromatic field of the H-NMR spectra complicated signal
groups proving the partial oligomerisation of the starting molecules. d Other products obtained from 2b and 2c will be reported below. e Traces of
product. f After 48 hours and supplementary addition of FeCl3, the triarylmethanol 9 R = OCH3 was formed in 6.5% yield only.
Besides the products described in Table 1, the oxidation of 2b
also yields 3,3Ј-dichloro-1,1Ј-biazulene, 12, (6%) and the oxid-
ation of 2c affords the products 13 (8.5%), 14 (20%) and 15
(2%).
Scheme 1
The general features of all products obtained are the similar-
1
ity in the H-NMR spectra of the azulene moiety proton sig-
nals, both in the starting compounds 2 or 3 and in the products,
as well as the lack of the proton signal for benzyl CH3 in the
products. For the products 6 X = CH3 or Cl, a singlet at 4.03
ppm (2 H) characteristic for the CH2 protons in diarylmethane
compounds (for diphenylmethane, 3.96 ppm)7 can be observed.
The products 7 X = CH3 or Cl and 10 show, at 5.63 ppm, a
singlet (1 H) characteristic for CH in triarylmethane derivatives
(for triphenylmethane 5.53 ppm).7 The chemical shifts for these
proton signals are not influenced by substituents on the azulene
Compound 13 shows in its IR spectrum a carbonyl group
band at 1650 cmϪ1 and the presence of a C᎐O function is also
᎐
1
confirmed by the H-NMR spectrum. Thus, for all studied
compounds, including 13, the AB spin systems for disubstituted
phenyl rings showed small changes in chemical shift for pro-
1
moiety. The ratio between the proton signals in the H-NMR
spectra, and also the mass spectra, are in agreement with the
proposed structures for 6, 7 and 10. Diphenylmethane, 4,
and triphenylmethane, 5, were also separated from almost all
reaction mixtures.
tons vicinal to the N᎐N bond (7.8–7.9 ppm). However, the two
᎐
protons ortho to the carbonyl carbon in 13 are strongly shielded
(about 0.5 ppm) in comparison with the compounds in which
this atom is sp3 hybridised, suggesting the involvement of this
carbon atom in a carbonyl bond. The structure assignment for
13 as azulene-1-azo(4Ј-benzoylbenzene) was also confirmed by
other proton signals, by the mass spectrum and by elemental
analysis.
1
The singlet at 5.91 ppm (1 H) in the H-NMR spectrum of
the product 8 X = OCH3 can be assigned to a proton character-
istic of a diarylmethanol (for benzhydrol 5.67 ppm) and the IR
spectrum indicates, indeed, the presence of an OH function
1
1
for 8. The chemical shifts of other proton signals in the H-
The proton chemical shifts in the H-NMR spectrum of 14
NMR spectrum and the ratio between them together with the
mass spectrum are consistent with the diarylmethanol structure
for 8.
are almost identical to those for alcohol 8 X = OCH3, excepting
the chemical shift for the CH benzylic singlet. The hydroxy
proton vanished and in the IR spectrum the OH signal was not
detected, suggesting an ether structure for 14. In the mass
spectrum of compound 14 the molecular ion (m/z = 718) is
not detectable, however, the abundance of [Mϩ ϩ 1] is 100%.
Another observation is the presence of the peak at 359 (40%)
representing M/2. These observations can suggest the tendency
of ether 14 for protonation and for transfer of two electrons
with the generation of a dication during the mass spectrum
recording, and also the low stability of the molecular ion. In the
1H-NMR spectrum of 14 all proton signals are split with J ca.
1 Hz (the split signals have the same ratio in their intensities),
although the mass spectrum and elemental analysis seem to
The IR spectra of compounds 9 X = CH3 or Cl show the OH
1
signal (3580 cmϪ1, m, in CH2Cl2) and in the H-NMR spectra
only aromatic (phenyl and azulenyl) proton signals are present.
The 1H-NMR spectra were also recorded in CD2Cl2 for accur-
ate phenyl signals integration and the results are in accord with
a triarylmethanol structure for 9. A triarylmethanol structure
was also found for compound 11. The same conclusion con-
cerning the structure of 9 and 11 can be drawn from their mass
spectra.
The 13C-NMR and elemental analysis for all mentioned
products are also consistent with the structure assignments.
990
J. Chem. Soc., Perkin Trans. 1, 2000, 989–994

View Article Online
indicate the presence of a single compound. These results are in
agreement with the formation of a racemic and meso mixture
for the ether 14 with two identical asymmetric carbon atoms.
The 1H- and 13C-NMR spectra of the side-product 15, together
with COSY and HETCOR experiments, are in agreement with
an interesting structure having the benzyl position substituted
by both the azulene moiety and phenyl. Structure assignment
for the product 15 is based on the data reported in the Experi-
mental section (the purification and complete characterisation
of 15 could not be accomplished due to the small amount of
product formed).
The structure of the products obtained in the reaction of 2
and 3 indicates that the reaction proceeds exclusively at the
benzyl position (in a subsequent reaction the C–azo bond can
be broken with the formation of diphenylmethane, 3, and tri-
phenylmethane, 4). (a) 3-Methoxy compound 2c reacted fastest,
(b) the overall yield of the separated product decreases in order
2c, 2a, 3 and 2b and in the same order the amount of resulting
tar increases and (c) while the oxidation of the compound 2c
affords only diarylmethane compounds (8 X = OCH3, 13 and
14), triarylmethane compounds (7 X = CH3 and Cl and 9
X = CH3 and Cl, or 10 and 11) are formed as major reaction
products starting from 2a and 2b or from 3.
“superacid”.2–5,8–10 The resulting radical 18 can react with
oxygen in the air (step c) yielding products with O-containing
functions at the benzyl carbon (alcohol, aldehyde, etc.) or has
the possibility to dimerise to ethane derivative 19 (step d).
Because, generally, the neutral radical is more available for a
subsequent single-electron oxidation than its parent com-
pound,11 the radical 18 could be oxidised to the corresponding
cation 16 (step e). The structural peculiarities that stabilise the
positive charge of the radical cation (for example as a tropylium
cation moiety) are expected to weaken its acidity and therefore
decrease its aptitude for proton loss.8 Therefore, the absence of
oxygenated products with an ArCH2 skeleton and of com-
pounds 19 can be explained by the shifting of the equilibrium
ϩ
(step b) toward 2 and by a higher rate of the single electron
᝽
oxidation (step e) as compared to the rates of steps c and d (the
reversibility of step d can also be considered). Although the
benzyl radical cation shows a strong preference for the hetero-
lytic pathway, the alternative path involving a hydrogen atom
transfer has sometimes been considered.8,10 For the azulenic
radical cation 2 ϩ, with a well stabilised charge, hydrogen atom
᝽
transfer (step f) could be a favourable alternative reaction route.
This pathway allows the direct generation of the stabilised
cation 16 and the nonpolar reaction medium seems to favour
homolytic C–H bond dissociation. From the possible resonance
structures for 16, only two, the tropylium and benzyl structures,
16(T) and 16(B), respectively, are illustrated because only these
In the previous study concerning the dimerisation of the
azulene-1-azoarenes1 we have explained the product generation
by the coupling of two radical cations obtained in an oxidative
single-electron transfer from the azo compound to FeCl3. We
suggest a similar oxidation of the compounds 2 or 3 to the
corresponding radical cations as the first step in the reported
reactions (Scheme 2, step a). However, the subsequent
contribute significantly to the stability and reactivity of the
formed ion. The attack of 16 towards the benzene (step g) with
the generation of diarylmethane product 6 is a well-known
electrophilic aromatic benzylation.
From the starting materials investigated, 2c displays the
lowest oxidation potential and, consequently, the best reactiv-
ity. While triarylmethane compounds (7, 9, 10 and 11) were
the major products in the oxidation of 2a, 2b and 3, only the
diarylmethanol, 8 and the derivatives of this alcohol (benzo-
phenone 13 and ether 14) were formed in the reaction of 2c.
These results seem to confirm the enhanced stability of carbo-
cation 16 X = OCH3 in comparison with 16 X = CH3 or Cl,
which reacts again with benzene to produce triarylmethane
derivatives. The methyl groups attached to the seven-membered
ϩ
ring in 3 contribute to the stability of the cationic species (2
᝽
and 2ϩ) and can promote the oxidation (step a). However, in
spite of the steric blocking of position 3 in azulene, this posi-
tion remains partly reactive and can produce polymerisation,
a reaction supported by the overall yield diminution.
Scheme 2
The oxygenated products can result from the radical reaction
with oxygen in the air and the nucleophilic attack of water on
the cationic intermediates. The assumption that the benzene
alkylation proceeds via a heterolytic mechanism suggests that
the generation of oxygenated compounds involves the same
cation intermediates that attack the benzene.
The reaction of the cationic intermediates with water upon
the decomposition of reaction mixture generates the alcohols
(step h). However, the presence of the benzophenone 13 seems
to indicate that at least some alcohol is generated before the
work-up of the reaction mixture (the reaction occurs without
protection from the air). Alcoholic compounds could be also
obtained from the hydrolysis of the corresponding chloro-
compounds generated as intermediates from the carbocations,
ϩ
dimerisation of radical cation 2 (or 3 ϩ) is hindered because
᝽
᝽
both its reactive positions 3 and 4Ј are blocked. Therefore, this
intermediate must react further by another pathway, a path-
way that yields phenyl-substituted products at the benzyl
carbon. Benzene alkylation, generally, is an electrophilic reac-
tion and, therefore, cations such as 16 and 17 can be postulated
as alkylating reagents.
However, for the benzyl oxidative substitution of radical
cations the transfer of the benzyl proton to a base present in the
reaction medium, in an acid–base equilibrium,2,8 is ordinarily
expected to be the favoured pathway (step b) because a
radical cation with a proton attached to carbon is often
J. Chem. Soc., Perkin Trans. 1, 2000, 989–994
991

View Article Online
however the first pathway seems more probable because in the
reaction mixtures halogenated compounds were never found,
not even in traces.
(t), 136.07 (t), 139.08 (q), 140.47 (q), 140.76 (t), 152.48 (q),
152.66 (q); m/z (70 eV) 276 (Mϩ, 3%), 275 (Mϩ Ϫ 1, 2.7), 261
(Mϩ Ϫ 15, 2), 157 (C10H6OCH3 , 19), 142 (13), 128 (4.5), 127
ϩ
ϩ
The generation of di- and triphenylmethane, 4 and 5 requires
that the C–azo bond break. There are two possible pathways for
this bond dissociation, a homolytic as well as heterolytic one
and two possible breaking positions, azulene–azo or phenyl–
azo. Unfortunately, the available experimental evidence does
not differentiate between these pathways (the azulene–azulene
coupled product 12 would be consistent with a dimerisation
of two azulene radicals produced in a homolytic azulene–azo
bond break).
In conclusion we have found a system in which the benzyl
hydrogen substitution by phenyl occurs to a good extent
and generates di- and triarylmethane compounds and their
corresponding oxygenated derivatives (alcohols, ether and
benzophenone).
(4.5), 115 (9), 114 (100), 91 (C7H7 , 51.5), 88 (22.5), 65 (47.5).
4,6,8-Trimethylazulene-1-azo(4Ј-methylbenzene), 3. Brown-
black microcrystalline powder, mp 155–156 ЊC (from ethyl
ether on precipitation with n-pentane) (Found: C, 83.37; H,
6.94; N, 9.69. C20H20N2 requires C, 83.30; H, 6.99; N, 9.71%);
λmax(dioxane)/nm 245 (log ε/dm3 molϪ1 cmϪ1 4.33) 292 (4.26),
325 (4.12), 440 (4.39); δH(CDCl3) 2.41 (3 H, s, 4Ј-CH3), 2.59
(3 H, s, 6-CH3), 2.83 (3 H, s, 4-CH3), 3.33 (3 H, s, 8-CH3), 7.12
(1 H, s, 5-H), 7.23 (1 H, s, 7-H), 7.27 (2 H, d, J 8.2, 3Ј-, 5Ј-H),
7.33 (1 H, d, J 4.5, 3-H), 7.77 (2 H, d, J 8.5, 2Ј-, 6Ј-H), 8.12 (1 H,
d, J 4.8, 2-H); δC(CDCl3) 21.41 (p), 25.30 (p), 28.54 (p), 29.64
(p), 117.83 (t), 122.01 (t), 122.28 (t), 129.62 (t), 130.38 (t),
132.67 (t), 133.40 (q), 138.82 (q), 140.31 (q), 146.95 (q), 147.41
(q), 147.50 (q), 149.61 (q), 152.49 (q); m/z (70 eV) 289 (Mϩ ϩ 1,
2%), 288 (Mϩ, 9.5), 287 (Mϩ Ϫ 1, 7.0), 273 (Mϩ Ϫ CH3, 13),
182 (100).
Experimental
Melting points: Kofler apparatus (Reichert, Austria). Elemental
analyses: Perkin Elmer CHN 240B. UV: Beckman DK-2A, UV
5240. 1H- and 13C-NMR: Bruker WM 300, AC 300, ARX 300,
and Gemini 300 (1H: 300 MHz, 13C: 75.47 MHz), J values are
given in Hz, TMS was used as internal standard in CDCl3 or
CD2Cl2 as solvent; signals were assigned on the basis of COSY
and HETCOR correlation spectra (the numbering for the com-
pounds was indicated on the structures and in Scheme 1);
abbreviations used for 13C-NMR: p, primary; s, secondary; t,
tertiary; q, quaternary. IR: Beckmann IR 5A. Mass spectra:
Finnigan MAT 311-A/100MS. Column chromatography: basic
alumina [activity BII-III (Brockmann)] or silica gel [70–230
mesh (ASTM)]. Dichloromethane (DCM) was distilled over
calcium hydride, ethyl acetate over anhydrous sodium carbon-
ate and trichloromethane was filtered through basic alumina.
Oxidation with ferric chloride–general procedure
In a round-bottomed flask, to a magnetically stirred solution of
1 mmol azocompound 2 or 3 in 25 cm3 benzene, anhydrous
ferric chloride was added at room temperature, without air
protection (the azocompound:FeCl3 molar ratio will be indi-
cated for each azocompound). From time to time the separated
viscous oily product was mechanically removed from the flask
wall for better contact between the reactants. The reaction
was monitored by TLC (homogenous samples were collected,
washed with a saturated sodium hydrogen carbonate solution
and dried) and it was stopped when the spot of the starting
compound had vanished or maintained a constant intensity
(the reaction time will be indicated for each azo compound).
The reaction mixture was quenched in a saturated sodium
hydrogen carbonate solution and the majority of the benzene
was evaporated in vacuo. The suspension was extracted three
times with 200 cm3 DCM, the organic extracts were washed
once with water, dried (Na2SO4) and the solvent was evaporated
in vacuo. The residue was separated by column chromatography
under the conditions described below.
Synthesis of starting azocompounds
For the synthesis of azulene-1-azoarenes we have adapted both
the Gerson and the Anderson protocols.6 The yields of the azo-
compounds were above 90% and the resulted compounds were
characterised. The synthesis protocol, the yield and the struc-
ture assignments for 2b have been reported in the literature.6c
Reaction of 3-methylazulene-1-azo(4Ј-methylbenzene), 2a.
The azocompound:FeCl3 ratio = 1:4; reaction time = 48 hours.
The resulting residue on work-up (290 mg) was separated by
chromatography on silica gel, eluent DCM–n-pentane (4:1);
fraction 1 (173 mg) was again separated on alumina; fraction 2
(4 mg), indefinite pink-coloured compound or mixture (with
complicated 1H-NMR spectrum and a proton signal above
10 ppm); fraction 3 (9 mg), complex mixture of components
with very close RF; fraction 4 (64 mg, 0.15 mmol) triarylmeth-
anol 9 X = CH3. Chromatography of fraction 1 was carried out
on alumina with n-pentane–ethyl acetate (50:1): fraction 1.1
(3 mg), green oily unidentified mixture; fraction 1.2 (7 mg)
mixture of diphenylmethane, 4, and triphenylmethane, 5,12 in
3-Methylazulene-1-azo(4Ј-methylbenzene), 2a. Black crystals,
mp 131–132 ЊC (n-hexane) (Found: C, 82.99; H, 6.23; N, 10.78.
C18H16N2 requires C, 83.04; H, 6.20; N, 10.76%); λmax(dioxane)/
nm 243 (log ε/dm3 molϪ1 cmϪ1 4.34), 285 (4.32), 338 (4.23), 437
(4.48), 600 sh (2.81); δH(CDCl3) 2.44 (3 H, s, 4Ј-CH3), 2.66 (3 H,
s, 3-CH3), 7.20 (1 H, t, J 9.8, 5-H), 7.26 (1 H, t, J 9.9, 7-H), 7.32
(2 H, d, J 8.3, 3Ј-, 5Ј-H), 7.64 (1 H, t, J 9.9, 6-H), 7.89 (2 H, dt,
J 8.3 and 1.7, 2Ј-, 6Ј-H), 8.15 (1H, s, 2-H), 8.20 (1 H, d, J 9.3,
4-H), 9.22 (1 H, d, J 9.5, 8-H); δC(CDCl3) 12.89 (p, 3-CH3),
21.57 (p, 4Ј-CH3), 122.20 (t), 125.21 (t), 125.42 (t), 125.73 (t),
128.68 (t), 129.78 (t), 134.99 (t), 135.14 (t), 139.37 (t), 141.37
(q), 142.60 (q), 152.42 (q); m/z (70 eV) 262 (Mϩ ϩ 2, 3%), 261
(Mϩ ϩ 1, 15), 260 (Mϩ, 74), 259 (Mϩ Ϫ 1, 46), 245 (Mϩ Ϫ CH3,
11), 141 (100), 115 (75).
1
molar ratio 1:1.5 (from H-NMR spectrum), yields 3% and
2%, respectively; fraction 1.3 (90 mg, 35 mmol) starting azo-
compound, 2a, and fraction 1.4 (66 mg), mixture of compounds
1
3-Methoxyazulene-1-azo(4Ј-methylbenzene) 2c. Black pow-
der, mp 102–103 ЊC (from CH2Cl2, on precipitation with
n-pentane) (Found: C, 78.12; H, 5.90; N, 10.16. C18H16N2O
requires C, 78.24; H, 5.84; N, 10.14%); λmax(dioxane)/nm 256
(log ε/dm3 molϪ1 cmϪ1 4.25), 292 (4.21), 350 (4.24), 472 (4.41),
493 (4.41); δH(CDCl3) 2.45 (3 H, s, CH3 in 4Ј), 4.05 (3 H, s,
OCH3), 6.88 (1 H, t, J 9.7, 5-H or 7-H), 6.97 (1 H, t, J 9.6, 7-H
or 5-H), 7.26 (2 H, dt, J 8.8 and 2.2, 3Ј-, 5Ј-H), 7.41 (1 H, t,
J 9.9, 6-H), 7.62 (1 H, s, 2-H), 7.85 (2 H, dt, J 8.6 and 2.1, 2Ј-,
6Ј-H), 8.15 (1 H, dd, J 9.1 and 1.1, 4-H), 9.02 (1-H, d, J 9.5,
8-H); δC(CDCl3) 21.52 (p), 57.77 (p), 102.57 (t), 122.09 (t),
124.51 (t), 124.58 (t), 129.75 (t), 131.46 (q), 134.09 (q), 134.25
6 X = CH3 and 7 X = CH3 molar ratio 1:1.45 (from H-NMR
spectrum). The product 7 X = CH3 was separated from the
fraction 1.4 by chromatography on alumina with n-pentane–
ethyl acetate (100:1).
3-Methylazulene-1-azo[4Ј-(diphenylmethyl)benzene], 7 X =
CH3. Yield 16%, black microcrystalline powder, mp 141–142 ЊC
(from DCM on precipitation with n-pentane) (Found: C, 87.19;
H, 5.92; N, 6.89. C30H24N2 requires C, 87.35; H, 5.86; N,
6.79%); δH(CD2Cl2) 2.66 (3 H, s, 3-CH3), 5.63 (1 H, s,
(C6H5)2CH-), 7.13–7.35 (14 H, m, 5-, 7-, 3Ј-, 5Ј-H and 2 C6H5),
7.64 (1 H, t, J 9.9, 6-H), 7.87 (2 H, dt, J 9.2 and 1.0, 2Ј-, 6Ј-H),
8.13 (1 H, s, 2-H), 8.21 (1 H, d, J 9.6, 4-H), 9.18 (1 H, d, J 9.8,
992
J. Chem. Soc., Perkin Trans. 1, 2000, 989–994

View Article Online
8-H); δC(CD2Cl2) 12.78 (p, 3-CH3), 55.77 (t, (C6H5)2CH-),
122.10 (t), 125.28 (t), 125.45 (t), 125.83 (t), 126.46 (t), 128.42 (t),
128.70 (q), 129.55 (t), 130.12 (t), 134.94 (t), 135.10 (t), 139.08
(q), 139.33 (t), 141.48 (q), 142.68 (q), 143.81 (q), 145.02 (q),
152.88 (q); m/z (70 eV) 414 (Mϩ ϩ 2, 2%), 413 (Mϩ ϩ 1, 8),
412 (Mϩ, 35), 411 (Mϩ Ϫ 1, 39), 397 (Mϩ Ϫ CH3, 3), 141 (100),
115 (92).
molϪ1 cmϪ1 4.48), 270 (4.67), 301 sh (4.45), 313 (4.54), 334 sh
(3.92), 395 (4.16), 638 (2.86); δH(CDCl3) 7.07 (2 H, t, J 9.8, 7-,
7Ј-H), 7.19 (2 H, t, J 9.7, 5-, 5Ј-H), 7.61 (2 H, t, J 9.8, 6-, 6Ј-H),
7.92 (2 H, s, 2-, 2Ј-H), 8.24 (2 H, d, J 9.4, 8-, 8Ј-H), 8.42 (2 H, d,
J 9.8, 4-, 4Ј-H); δC(CDCl3) 76.59 (q), 77.02 (q), 77.44 (q), 116.53
(q), 123.13 (t, C-5), 123.74 (t, C-7), 135.00 (t, C-4), 135.79 (t,
C-2), 136.96 (t, C-8), 139.79 (t, C-6); m/z (70 eV) 327 (Mϩ ϩ 4,
3%), 326 (Mϩ ϩ 3, 13), 325 (Mϩ ϩ 2, 16), 324 (Mϩ ϩ 1, 77), 323
(Mϩ, 26), 322 (Mϩ Ϫ 1, 100), 161 (13).
3-Methylazulene-1-azo[4Ј-(phenylmethyl)benzene],
6
X =
CH3. Yield 11% (from 1H-NMR spectrum), contaminated with
7 X = CH3, black powder; δH (CD2Cl2) 2.63 (3 H, s, 3-CH3), 4.03
(0.8 H, s, (C6H5)CH2- in 6), 5.63 (0.6 H, s, (C6H5)2CH- in 7); the
signals in the aromatic field are common for 6 and 7: 7.15–7.35
(m, 3Ј-, 5Ј-, 7-, 5-H and C6H5), 7.65 (1 H, t, J 9.9, 6-H), 7.86
(2 H, dt, J 8.8 and 1.9 Hz, 2Ј-, 6Ј-H), 8.10 (1 H, s, 2-H), 8.21
(1 H, d, J 9.3, 4-H), 9.17 (1 H, d, J 9.8, 8-H); m/z (FD) 415
(Mϩ ϩ 3, 4%), 414 (Mϩ ϩ 2, 8), 413 (Mϩ ϩ 1, 100), 412 (Mϩ,
15) for 7 and 339 (Mϩ ϩ 3, 2), 338 (Mϩ ϩ 2, 5), 337 (Mϩ ϩ 1,
36), 336 (Mϩ, 2) for 6.
3-Chloroazulene-1-azo[4Ј-(diphenylmethyl)benzene],
7
X = Cl, in a mixture with 3-chloroazulene-1-azo[4Ј-(phenyl-
methyl)benzene], 6 X = Cl. Black powder, δH(CDCl3) 4.05 (0.30
H, s, (C6H5)CH2- in 6 X = Cl), 5.62 (0.85 H, s, (C6H5)2CH- in 7
X = Cl), the protons signals in aromatic field for 6 and 7 are
common: 7.13–7.41 (12 H, m, 3-, 7-H and phenyl protons), 7.86
(2-H, dt, J 8.5 and 2.0, 2Ј-, 6Ј-H), 8.19 (1 H, s, 2-H), 8.41 (1 H,
d, J 9.4, 4-H), 9.24 (1 H, d, J 9.9, 8-H); δC(CDCl3) 56.71 (s),
120.98 (t), 122.22 (t), 122.86 (t), 126.20 (t), 126.44 (t), 126.65
(q), 128.38 (t), 128.52 (t), 128.69 (t), 129.46 (t), 130.11 (q),
136.05 (t), 136.18 (q), 137.41 (q), 140.83 (q), 143.60 (t), 145.68
(q), 152.40 (q); m/z (15 eV) 434/432 (M(7)ϩ, 35/100%), 435/433
(M(7)ϩ ϩ 1, 8.5/37), 431 (39), 397 (M(7)ϩ Ϫ 35, 8.5), 379
(14.5%), 378/376 (M(6)ϩ, 60/100), 377 (45), 375 (41), 358 (12),
356 (43), 341 (10), 189 (11.5), 166 (12), 163 (28.5), 161 (82).
3-Chloroazulene-1-azo[4Ј-(diphenylhydroxymethyl)benzene],
9 X = Cl. Yield 22%, black powder, mp above 260 ЊC (at 250 ЊC
slight decomp., from DCM on precipitation with n-pentane)
(Found: C, 77.30; H, 4.85; N, 6.39; Cl, 7.90. C29H21ClN2O
requires C, 77.59; H, 4.71; N, 6.24; Cl, 7.89%); λmax(dioxane)/
nm 287 (log ε/dm3 molϪ1 cmϪ1 4.32), 336 (4.26), 436 (4.47), 455
sh (4.40), 600 sh (2.96), 675 sh (2.76); νmax (CH2Cl2)/cmϪ1 3580
m (OH); δH(CDCl3) 2.90 (1 H, s, OH), 7.25–7.38 (12 H, m, 5-,
7-H and 2 C6H5), 7.42 (2 H, dt, J 8.7 and 2.0, 3Ј-, 5Ј-H), 7.74
(1 H, t, J 9.8, 6-H), 7.87 (2 H, dt, J 8.7 and 2.0, 2Ј-, 6Ј-H), 8.19
(1 H, s, 2-H), 8.40 (1 H, d, J 9.3, 4-H), 9.23 (1 H, d, J 9.6, 8-H);
δC(CDCl3) 82.01 (q), 121.75 (t), 122.84 (t), 126.33 (t), 126.80 (t),
127.38 (t), 127.91 (t), 128.00 (t), 128.62 (t), 135.30 (q), 136.03
(t), 136.20 (t), 137.26 (q), 137.52 (q), 140.86 (t), 141.19 (q),
146.65 (q), 148.14 (q), 152.84 (q); m/z (70 eV) 450/448 (Mϩ, 25/
75%), 160 (100).
3-Methylazulene-1-azo[4Ј-(diphenylhydroxymethyl)benzene],
9 X = CH3. Yield 20%, brown-black crystals, mp 123–124 ЊC
(from DCM on precipitation with n-pentane) (Found: C,
83.10; H, 5.71; N, 6.22. C30H24N2O requires C, 83.07; H, 5.65;
N, 6.54%; λmax(dioxane)/nm 240 (log ε/dm3 molϪ1 cmϪ1 4.35),
286 (4.30), 340 (4.42), 441 (4.50), 600 sh (2.88); νmax(CH2Cl2)/
cmϪ1 3580 (OH); δH(CD2Cl2) 2.66 (3 H, s, 3-CH3), 2.98 (1 H, s,
OH), 7.28 (1 H, t, J 9.7, 5-H), 7.27–7.36 (11 H, m, 7-H and
2 C6H5), 7.32 (2 H, dt, J 8.6 and 1.8, 3Ј-, 5Ј-H), 7.70 (1 H, t,
J 9.9, 6-H), 7.88 (2 H, dt, J 8.8 and 1.9, 2Ј-, 6Ј-H), 8.11 (1 H, s,
2-H), 8.27 (1 H, d, J 9.4, 4-H), 9.20 (1 H, d, J 9.5, 8-H);
δC (CD2Cl2) 13.27 (p, 3-CH3), 86.62 (q), 122.23 (t), 125.85 (t),
126.41 (t), 126.91 (t), 128.11 (t), 128.66 (t), 128.77 (t), 129.40 (t),
129.80 (q), 135.57 (t), 136.12 (t), 140.02 (q), 140.33 (t),
142.35 (q), 143.33 (q), 147.65 (q), 148.44 (q), 154.05 (q); m/z
(FD) 430 (Mϩ ϩ 2, 7.5%), 429 (Mϩ ϩ 1, 42), 428 (Mϩ, 100);
m/z (70 eV) 430 (Mϩ ϩ 2, 4%), 429 (Mϩ ϩ 1, 20), 428 (Mϩ, 65),
427 (Mϩ Ϫ 1, 11), 141 (100), 115 (46).
Reaction of 3-chloroazulene-1-azo(4Ј-methylbenzene), 2b.
Ferric chloride was added in two portions, 4 mmol at the begin-
ning of the stirring and 2 mmol after three hours (addition in
only one portion lowers the yield); the total reaction time was
48 hours. The resulting residue on work-up (340 mg) was separ-
ated on silica gel, eluent n-pentane–ethyl acetate (from 20:1 to
1:1): fraction 1 (150 mg) and fraction 2 (12 mg) were again
chromatographed; fraction 3 (23 mg), was a complex unidenti-
fied mixture; fraction 4 (59 mg, 0.13 mmol), triarylmethanol 9
X = Cl; fraction 5 (20 mg), a mixture of oligomers with complex
Reaction of 3-methoxyazulene-1-azo(4Ј-methylbenzene), 2c.
The azocompound:FeCl3 ratio = 1:4; the reaction time = 4
hours. The resulting residue on work-up (320 mg) was separ-
ated on silica gel with n-pentane–ethyl acetate (5:1): fraction
1 (3 mg) in the 1H-NMR spectrum shows the signals of
diphenylmethane, 4, and triphenylmethane, 5;12 fraction 2 (40
mg, 0.15 mmol), starting compound, 2c; fraction 3 (27 mg,
0.074 mmol), ketone 13; fraction 4 (62 mg, 0.086 mmol), ether
14; fraction 5 (12 mg) was again chromatographed; fraction 6
(94 mg, 0.26 mmol), diarylmethanol 8 X = OCH3; fraction 5
was separated on silica gel, eluent DCM–n-pentane (3:1):
fraction 5.1 (4 mg, 0.008 mmol), compound 15 and fraction 5.2
(6 mg), unidentified mixture.
3-Methoxyazulene-1-azo[4Ј-(phenylhydroxymethyl)benzene],
8 X = OCH3. Yield 30%, black powder, mp 59–61 ЊC (from
DCM on precipitation with n-pentane) (Found: C, 78.93;
H, 5.38; N, 7.51. C24H20N2O2 requires C, 78.24; H, 5.47; N,
7.60%); λmax(dioxane)/nm 250 (log ε/dm3 molϪ1 cmϪ1 4.42), 292
(4.35), 350 (4.36), 471 (4.41), 492 (4.41); νmax(CH2Cl2)/cmϪ1
3585 m (OH); δH(CDCl3) 4.06 (3 H, s, OCH3), 5.91 (1 H, s, CH
in 4Ј), 7.02 (1 H, t, J 9.6, 5- or 7-H), 7.09 (1 H, t, J 9.9, 7- or
5-H), 7.23–7.28 (1 H, m, 4Љ-H), 7.35 (2 H, t, J 7.6, 3Љ-, 5Љ-H),
7.42 (2 H, d, J 7.2, 2Љ-, 6Љ-H), 7.49 (2 H, dt, J 8.5 and 1.7, 3Ј-,
5Ј-H), 7.54 (1 H, t, J 9.9, 6-H), 7.65 (1 H, s, 2-H), 7.90 (2 H,
dt, J 8.6 and 2.0, 2Ј-, 6Ј-H), 8.25 (1 H, dd, J 9.1, and 1.3,
4-H), 9.08 (1 H, d, J 9.9, 8-H); δC(CDCl3) 57.73 (p), 75.94 (t),
102.43 (t), 122.04 (t), 124.77 (t), 124.88 (t), 126.62 (t), 127.17
(t), 127.55 (t), 128.47 (t), 131.68 (q), 134.33 (t), 134.51 (q),
136.03 (t), 140.45 (q), 140.81 (t), 143.80 (q), 144.39 (q),
1
signals in the aromatic field of H-NMR spectrum. Fraction 1
was separated on silica gel with n-pentane: fraction 1.1 (6 mg,
0.02 mmol) 3,3Ј-dichloro-1,1Ј-biazulene, 12; fraction 1.2 (25.5
mg), mixture of diphenylmethane, 4, and triphenylmethane, 5,12
molar ratio 1:3 (from 1H-NMR), yields 4.5% and 13%, respec-
tively; fraction 1.3 (6.5 mg), uncharacterised complex mixture;
fraction 1.4 (112 mg, 0.4 mmol) starting material, 2b. Fraction 2
was separated on silica gel with n-pentane–DCM 2:1: fraction
2.1 (4.5 mg), uncharacterised mixture, fraction 2.2 (6 mg),
1
mixture of 6 X = Cl, and 7 X = Cl, molar ratio 1:4 (from H-
NMR), yields 2% and >1%, respectively, and fraction 2.3 (1.5
mg), complex unidentified mixture. It cannot be claimed that
analysis of the spectra of fraction 2.2 allows unequivocal struc-
ture assignment for the compounds 6 X = Cl and 7 X = Cl.
However, the chemical shifts of many signals in the recorded
spectra of fraction 2.2 are similar to those present in the spectra
of other compounds 6 and 7 (for example the aliphatic proton
signals for C6H5CH2- or for (C6H5)2CH-). GCMS is not useful
in this case because of the very low volatility of the products.
3,3Ј-Dichloro-1,1Ј-biazulene, 12. Yield 6%, green needles, mp
143–144 ЊC (from diethyl ether on precipitation with n-pentane)
(Found: C, 74.30; H, 3.71; Cl, 21.99. C20H12Cl2 requires C,
74.32: H, 3.74; Cl, 21.94%); λmax(dioxane)/nm 246 sh (log ε/dm3
J. Chem. Soc., Perkin Trans. 1, 2000, 989–994
993

View Article Online
152.74 (q), 153.68 (q); m/z (25 eV) 370 (Mϩ ϩ 2, 2.5%), 369
(Mϩ ϩ 1, 19), 368 (Mϩ, 88), 367 (Mϩ Ϫ 1, 80), 353 (Mϩ Ϫ 15,
2), 352 (52.5), 351 (51), 337 (12), 218 (9), 165 (20), 157 (100),
143 (14), 142 (37), 137 (22), 136 (24), 135 (12), 123 (17), 121
(24), 115 (10), 114 (59).
(from trichloromethane on precipitation with n-pentane)
(Found: C, 87.30; H, 6.41; N, 6.29. C32H28N2 requires C, 87.22;
H, 6.42; N, 6.36%); λmax(dioxane) 245 nm (log ε/dm3 molϪ1 cmϪ1
4.42), 295 (4.16), 328 (4.19), 442 (4.49); δH(CDCl3) 2.61 (s, 3 H,
CH3), 2.85 (3 H, s, CH3), 3.32 (3 H, s, CH3), 5.61 (1 H, s,
(C6H5)2CH-), 7.15 (1 H, s, 5-H or 7-H), 7.17 (4 H, d, J 7.0, 2Љ-,
6Љ-H), 7.23 (2 H, d, J 8.3, 3Ј-, 5Ј-H), 7.26 (1 H, s, 7-H or 5-H),
7.20–7.30 (2 H, m, 4Љ-H), 7.31 (4 H, t, J 7.4, 3Љ-, 5Љ-H), 7.34
(1 H, d, J 5.0, 3-H), 7.80 (2 H, d, J 8.5, 2Ј-, 6Ј-H), 8.12 (1 H, d,
J 4.7, 2-H); δC(CDCl3) 25.33 (p, CH3), 28.54 (p, CH3), 29.64
(p, CH3), 56.78 (t), 117.95 (t), 122.04 (t), 122.28 (t), 126.42 (t),
128.39 (t), 129.53 (t), 130.06 (t), 130.56 (t), 132.87 (t), 133.60
(q), 140.49 (q), 143.86 (q), 144.60 (q), 147.06 (q), 147.62 (q),
149.67 (q), 152.87 (q); m/z (70 eV) 441 (Mϩ ϩ 1, 4.5%), 440
(Mϩ, 11), 439 (Mϩ Ϫ 1, 10), 425 (Mϩ Ϫ CH3, 15.5), 182 (100),
181 (73.5).
4-(3Љ-Methoxyazulene-1Љ-azo)benzophenone, 13. Yield 8.5%,
black powder, mp 147 ЊC (from ether on precipitation with
n-pentane) (Found: C, 78.03; H, 4.90; N, 7.55. C24H18N2O
requires C, 78.67; H, 4.95; N, 7.64%); λmax(dioxane) 245 nm (log
ε/dm3 molϪ1 cmϪ1 4.17), 290 (4.06), 356 (3.98), 500 (4.14), 690 sh
(2.71); νmax(CCl4)/cmϪ1 1650 m (C᎐O); δ (CDCl ) 4.08 (3 H, s,
᎐
H
3
OCH3), 7.11 (1 H, t, J 9.9, 7Љ-H or 5Љ-H), 7.20 (1 H, t, J 9.9,
5Љ-H or 7Љ-H), 7.48–7.58 (3 H, m, 3-, 4-, 5-H), 7.61 (1 H, t, J 9.6,
6Љ-H), 7.66 (1 H, s, 2Љ-H), 7.83 (2 H, d, J 7.7, 2-, 6 -H), 7.93 (2 H,
dAB, J 8.6, 3Ј-, 5Ј-H or 2Ј-, 6Ј-H), 7.96 (2 H, dAB, J 8.6, 2Ј-, 6Ј-H
or 3Ј-, 5Ј-H), 8.3 (1 H, d, J 9.4, 4Љ-H), 9.14 (1 H, d, J 9.9, 8Љ-H);
m/z (15 eV) 369 (Mϩ ϩ 3, 2%), 368 (Mϩ ϩ 2, 8), 367 (Mϩ ϩ 1,
26), 366 (Mϩ, 100), 365 (Mϩ Ϫ 1, 67), 351 (Mϩ Ϫ 15, 14), 341
(4), 157 (52), 136 (68), 123 (40), 121 (44), 109 (34).
4,6,8-Trimethylazulene-1-azo[4Ј-(diphenylhydroxymethyl)-
benzene], 11. Yield 17%, brown crystalline powder, mp 144–
146 ЊC (from trichloromethane on precipitation with n-pentane)
(Found: C, 84.25; H, 6.14; N, 6.10. C31H28N2O requires C,
84.17; H, 6.19; N, 6.12%); λmax(dioxane) 246 nm (log ε/dm3
meso-
and
(RS)-Bis[(3-methoxyazulene-1-azophenyl)-
(phenyl)methyl] ether, 14. Yield 20%, black powder (not defi-
nite mp) (Found C, 79.97; H, 5.41; N, 7.70. C24H38N4O3 requires
C, 80.20; H, 5.33; N, 7.79; λmax(dioxane)/nm 290 (log ε/dm3
molϪ1 cmϪ1 4.11), 292 (4.03), 349 (3.97), 487 (4.12); δH(CDCl3)
4.050 and 4.055 (3 H, 2 s, integral ratio ca. 0.7:2.3, OCH3), 5.51
(1 H, s, OCH), 6.99 and 7.00 (1 H, 2 t, J 9.9, 7-H or 5-H), 7.06
and 7.08 (1 H, 2 t, J 9.9, 5-H or 7-H), 7.24 and 7.31 (1H, m, 4Љ-
H), 7.34 and 7.35 (2 H, 2 t, J 7.0, 3Љ-, 5Љ-H), 7.43 (2 H, d, J 7.1,
2Љ-, 6Љ -H), 7.49 (2 H, d, J 8.3, 3Ј-, 5Ј-H), 7.52 and 7.53 (1 H, 2 t,
J 9.9, 6-H), 7.647 and 7.654 (1 H, 2 s, 2-H), 7.89 and 7.90 (2 H,
2 d, J 8.3, 2Ј-, 6Ј-H), 8.23 and 8.24 (1 H, 2 d, J 9.8, 4-H), 9.07
and 9.08 (1 H, 2 d, J 9.9, 8-H); δC(CDCl3) 57.75 (p, OCH3),
79.98 (t, O-CH), 122.07 (t), 124.51 (t), 124.56 (t), 124.73 (t),
124.82 (t), 124.85 (t), 127.29 (t), 127.34 (t), 127.96 (q), 128.01
(t), 128.41 (t), 128.44 (t), 131.69 br (q), 134.31 (t), 134.50 br (q),
136.07 (t), 140.50 (q), 140.52 (q), 140.80 (t), 142.07 (q), 142.61
(q), 142.71 (q), 142.98 (q), 152.74 br (q), 153.79 (q), 153.82
(q); m/z (FD): (Mϩ ϩ 3, 33%), (Mϩ ϩ 2, 83), (Mϩ ϩ 1, 100),
(M2ϩ, 40).
molϪ1 cmϪ1 4.31), 293 (4.06), 324 (4.07), 440 (4.33); νmax
-
(CH2Cl2) 3580 cmϪ1 m (OH); δH(CDCl3) 8.12 (2 H, d, J 5.0,
2-H), 7.80 (2 H, d, J 8.6, 2Ј-, 6Ј-H), 7.39 (2 H, d, J 8.6, 3Ј-,
5Ј-H), 7.27–7.35 (10 H, m, 2 C6H5), 3.32 (3 H, s, CH3), 2.85
(3 H, s, CH3), 2.61 (3 H, s, CH3); δC(CDCl3) 25.33 (p, CH3),
28.54 (p, CH3), 29.64 (p, CH3), 82.07 (q, C-OH), 118.06 (t),
121.81 (t), 122.05 (t), 127.35 (t), 128.00 (t), 128.65 (t), 130.73 (t),
133.04 (t), 133.83 (q), 140.67 (q), 146.88 (q), 147.17 (q), 147.55
(q), 147.70 (q), 148.93 (q), 149.72 (q), 153.36 (q); m/z (70 eV)
457 (Mϩ ϩ 1, 3%), 456 (Mϩ, 8.4), 455 (Mϩ Ϫ 1, 13), 441 (Mϩ Ϫ
CH3, 18.7), 182 (89.0), 181 (100).
Acknowledgements
A. C. R. thanks the A. v. Humboldt Foundation for a research
fellowship and express his gratitude to Professor Klaus Hafner
from T. U. Darmstadt for helpful suggestions and discussions.
This investigation was partly supported by the National Agency
of Science, Technology and Innovations–Bucharest.
[4Ј-(3Љ-Methoxyazulene-1Љ-azo)phenyl](3ٞ-methoxyazulen-
yl)phenylmethane, 15. Yield 2%, black powder; δH(CDCl3) 3.94
(3 H, s, OCH3), 4.08 (3 H, s, OCH3), 6.19 (1 H, s, C6H5CH),
6.67 (1 H, t, J 9.6, 7ٞ-H or 5ٞ-H), 6.75 (1 H, t, J 9.6, 5ٞ-H
or 7ٞ-H), 7.02 (1 H, t, J 9.7, 5Љ-H), 7.04 (1 H, s, 2ٞ-H), 7.08
References
1 A. C. Razus, J. Chem. Soc., Perkin Trans. 1, preceding paper DOI:
10.1039/a907402c.
2 V. D. Parker, Y. T. Chao and G. Zheng, J. Am. Chem. Soc., 1997,
119, 11390 and references cited therein.
3 D. M. Camanioni and J. A. Franz, J. Org. Chem., 1984, 49, 1607.
4 C. Walling, D. M. Camanioni and S. S. Kim, J. Am. Chem. Soc.,
1978, 100, 4814.
5 C. J. Schlesener, C. Amatore and J. K. Kochi, J. Am. Chem. Soc.,
1984, 106, 3567; L. M. Tolbert and R. K. Khanna, J. Am. Chem.
Soc., 1987, 109, 3477; L. M. Tolbert, R. K. Khanna, A. E. Popp,
L. Gelbaum and L. A. Bottomley, J. Am. Chem. Soc., 1990, 112,
2373.
(1 H, t, J 9.7, 7Љ-H), 7.19 (2 H, d, J 7.2, 2-, 6-H), 7.26 (2 H, dAB
,
J 8.4, 2Ј-, 6Ј-H), 7.27–7.32 (3 H, m, 3-, 4-, 5-H), 7.34 (1 H, t,
J 10, 6ٞ-H), 7.54 (1 H, t, J 10.0, 6Љ-H), 7.67 (1 H, s, 2Љ-H),
7.87 (2 H, dAB, J 8.4, 3Ј-, 5Ј-H), 8.04 (1 H, d, J 9.6, 4ٞ-H or
8ٞ-H), 8.19 (1 H, d, J 9.6, 8ٞ-H or 4ٞ-H), 8.25 (1 H, d, J 9.2,
4Љ-H), 9.08 (1 H, d, J 9.6, 8Љ-H).
Reaction of 4,6,8-trimethylazulene-1-azo(4Ј-methylbenzene),
3. The ratio azocompound:FeCl3 = 1:4; the reaction time 48
hours. The resulting residue on work-up (267 mg), was chrom-
atographed on silica gel with DCM–ethyl acetate (from 1:0 to
100:1): fraction 1 (15 mg), diphenylmethane 4 and triphenyl-
methane 5,12 molar ratio 0.15:1; fraction 2 (150 mg) was once
more separated; fraction 3 (75 mg) was, also, again separated;
fraction 4 (12 mg) mixture of oligomers. Fraction 2 was
chromatographed on alumina with n-pentane–ethyl ether (from
20:1 to 1:1): fraction 2.1 (59 mg), mixture of 4, 5 and starting
6 (a) F. Gerson, J. Schulze and E. Heilbronner, Helv. Chim. Acta,
1958, 41, 1444; (b) A. G. Anderson, Jr., J. A. Nelson and J. J.
Tazuma, J. Am. Chem. Soc., 1953, 75, 4980; (c) A. C. Razus, L.
Birzan, S. A. Razus and V. Horga, Rev. Roum. Chim., 1999, 44(3),
235.
7 C. J. Pouchert and J. Behnke, The Aldrich Library of 13C and 1H
FT-NMR Spectra, 19, edition I, vol. 2.
8 X. Zhang and F. C. Bordwell, J. Org. Chem., 1992, 57, 4163.
9 E. Baciocchi, T. Del Giacco and F. Elisei, J. Am. Chem. Soc., 1993,
115, 12290.
10 E. Baciocchi, T. Del Giacco, F. Elisei and O. Lanzalunga, J. Am.
Chem. Soc., 1998, 120, 11800 and the references cited therein.
11 L. Eberson, Electron Transfer Reactions in Organic Chemistry,
Springer Verlag, Berlin, 1987, p. 111.
12 The hydrocarbons 4 and 5 were separated from the eluted mixture
and their physical characteristics were identical with those of
diphenylmethane and triphenylmethane, respectively.
1
material 3, molar ratio 1:2:12.5 (molar, from H-NMR sig-
nals), fraction 2.2, 33 mg, mixture of 3 and triarylmethane 10,
1
ratio 1.8:1 (from H-NMR) and fraction 2.3 (56 mg), triaryl-
methane 10. Fraction 3 was chromatographed on silica gel,
eluent DCM–ethyl acetate (40:1): fraction 3.1, 12 mg
complex unidentified mixture and fraction 3.2 (60 mg), triaryl-
methanol 11.
4,6,8-Trimethylazulene-1-azo[4Ј-(diphenylmethyl)benzene],
10. Yield 21%, brown microcrystalline powder, mp 208–209 ЊC
Paper a908862h
994
J. Chem. Soc., Perkin Trans. 1, 2000, 989–994