Reactions of 1-(2-alkoxyphenyl)alkaniminyl radicals

Article abstract of DOI:10.1039/b103181n

Reactions of 1-(2-alkoxyphenyl)alkaniminyl radicals were studied. Generation of the iminyls 1d and 1e by standard methods in the gas-phase by flash vacuum pyrolysis (FVP) gives rise to 1,3-benzoxazines 18 and 22 by a mechanism which probably involves a 1,6-hydrogen atom translocation. The results showed that the dealkylation is an integral part of the reaction mechanism and not a subsequent reaction of 2-alkoxybenzonitriles.

Full text of DOI:10.1039/b103181n

View Article Online / Journal Homepage / Table of Contents for this issue
Reactions of 1-(2-alkoxyphenyl)alkaniminyl radicals
Rino Leardini,a Hamish McNab,*b Matteo Minozzi,a Daniele Nanni,*a David Reedb and
Andrew G. Wrightb
1
a Dipartimento di Chimica Organica “A Mangini”, Università di Bologna,
Viale Risorgimento 4, I-40136, Bologna, Italy
b Department of Chemistry, The University of Edinburgh, West Mains Road, Edinburgh,
UK EH9 3JJ
Received (in Cambridge, UK) 9th April 2001, Accepted 22nd August 2001
First published as an Advance Article on the web 27th September 2001
Genereration of the iminyls 1d and 1e by standard methods in the gas-phase by flash vacuum pyrolysis (FVP), or
in solution, respectively, gives rise to 1,3-benzoxazines 18 and 22 by a mechanism which probably involves a 1,6-
hydrogen atom translocation. Under FVP conditions, 2-cyanophenol 14 is a major product from 2-alkoxyphenyl-
alkaniminyls 1b–d; the dealkylation is an integral part of the reaction mechanism and not a subsequent reaction
of 2-alkoxybenzonitriles, which themselves are only very minor products of the pyrolyses.
If necessary the alkoxybenzaldehyde was first obtained from
salicylaldehyde 4a by treatment with the appropriate alkyl
Introduction
We have previously examined the cyclisation reactions of
iminyl radicals 1 bearing ortho-phenoxy,1,2 anilino,1,3 and thio-
phenoxy4,5 substituents, generating the intermediates both in
solution-phase1,4 and in the gas-phase under flash vacuum
pyrolysis (FVP) conditions.2,3,5 In the reactions of the N-methyl-
anilino intermediates, products were surprisingly obtained by
mechanisms in which the iminyl had clearly interacted with
the N-alkyl substituent rather than with the aromatic ring.1,3
For example, gas-phase generation of the iminyl 1a by FVP of
the oxime ether 2 led to the dihydroquinazoline 3 and the
mechanism shown in Scheme 1 was proposed to account for
halide in DMF containing potassium carbonate (see Scheme 2
Scheme 1
the transformation.3 In order to probe the generality of this
unexpected behaviour, we have explored the interaction of
iminyls 1b–e with ortho-alkoxy substituents and report the
results of both our solution-phase and gas-phase studies in this
paper.
Scheme 2 Reagents and conditions: (i) RBr, K2CO3, DMF; (ii)
MeONH2ؒHCl, pyridine, EtOH; (iii) HONH2ؒHCl, pyridine, EtOH;
(iv) ClCH2CO2H; (v) ButOOH, CDI.
Results
and Experimental section). The solution-phase precursor 13
was made by analogous alkylation of ketone 4b to the known
alkoxy derivative 7b, followed by reaction with hydroxylamine
hydrochloride. Following a standard procedure,7 the resulting
For the gas-phase work, the oxime ether precursors 8–10 were
made in standard fashion6 by reaction of the appropriate 2-
alkoxybenzaldehyde 5, 6 and 7a with O-methylhydroxylamine.
2704
J. Chem. Soc., Perkin Trans. 1, 2001, 2704–2710
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b103181n

View Article Online
oxime 11 was alkylated with chloroacetic acid to give com-
pound 12, which was finally transformed into perester 13 by
reaction with tert-butyl hydroperoxide in the presence of 1,1Ј-
carbonyldiimidazole (CDI) (Scheme 2).
ditions, which provided only 2-cyanophenol 14 (34%) and 2-
isopropoxybenzonitrile 17 (10%). The latter product has been
reported only once in the literature10 and so an authentic
sample was synthesised and its full characterisation is reported
in the Experimental section. In particular, the base peak in its
electron impact mass spectrum appears at m/z 119 due to loss of
propene from the molecular ion, presumably by a McLafferty-
type rearrangement involving the aromatic ring.
GC-MS analysis of the pyrolysate confirmed the presence
of 4a (m/z 132), 14 (m/z 119) and 17 (m/z 161) and revealed
one other major peak which had its molecular ion at m/z 161,
corresponding to loss of a hydrogen atom from the iminyl 1d.
This behaviour is reminiscent of the dihydroquinazoline for-
mation (Scheme 1) and a similar mechanism would give rise to
the 1,3-benzoxazine 18. The imine and C-dimethyl groups of
this structure would conveniently explain the signals at δH 8.08
and 1.53 respectively in the NMR spectrum of the crude
pyrolysate; in particular, the known11 2,2-diaryl derivatives
19–21 (in which the 2,2-substituents might be expected to have
a deshielding effect on the adjacent imine position) show imine
signals in the range δH 8.35–8.41. In addition, the base peak of
the mass spectrum was formed by a loss of 15 Da (= CH3) from
the molecular ion, which would be expected from a structure
such as 18.
FVP of the oxime ether 8 at 650 ЊC (ca. 0.01 Torr) gave a
1
complex pyrolysate, the H NMR spectrum of which showed
a surprising lack of significant resonances due to methoxy
groups. The presence of an aldehyde signal at δH 9.89 was
shown to be due to salicylaldehyde 4a (ca. 24%) by comparison
of the 13C NMR spectrum of the pyrolysate with literature
data.8 Dry-flash chromatography of the mixture gave only one
clean product which was identified as 2-cyanophenol 14 (46%)
by comparison of its spectrum with that of an authentic sample
(see Experimental section). All the major peaks in the aromatic
region of the 13C NMR spectrum of the crude pyrolysate are
accounted for by products 4a and 14 (Scheme 3). However, a
significant product which gave rise to a CH2 signal at δC 74.04
but which was not associated with any other peaks of com-
parable intensity, remains unidentified; it may be derived from
methoxyl radicals, formed as co-products in the generation of
the iminyl.
Nitriles are well-known decomposition products of iminyl
radicals, but, significantly, only a trace of 2-methoxybenzo-
nitrile 15 (ca. 5%) could be detected in the crude pyrolysate
from 8. Methoxy groups are also known to degrade to phenols
under FVP conditions, but generally at higher temperatures
than we have employed for iminyl generation.9 Indeed, control
pyrolyses of authentic 15 showed little evidence of dealkylation
at temperatures up to 750 ЊC and even at 850 ЊC ca. 30% of the
starting material remained. These results demonstrate that the
dealkylation process observed in the pyrolysis of 8 which gives
rise to 2-cyanophenol 14 is an integral part of the mechanism
and not merely a secondary decomposition.
Pyrolysis of the ethoxy compound 9 under the same condi-
tions as those used for the methoxy analogue 8 also gave rise to
2-cyanophenol 14 and salicylaldehyde 4a in a ca. 3 : 1 ratio as
the major products (Scheme 3). The assignment was clear from
the 13C NMR spectrum of the pyrolysate, by analogy with the
previous results (see Experimental section), and no further
analysis was carried out. Once again, compounds containing
alkyl groups (including the presence of a small amount of 2-
ethoxybenzonitrile 16) accounted for only a small proportion
of the products; the unidentified signal (δC 73.97) found in the
pyrolysis of 8 was again present.
The constitution of the 1,3-benzoxazine 18 was proved, as a
1 : 1 mixture with 2-cyanophenol 14 in the crude pyrolysate,
by the following sequence of NMR experiments. These were
carried out in [2H6]acetone solvent, in which seven of the eight
aromatic resonances were clearly resolved. Comparison with a
spectrum of 14 allowed the identification of the four aromatic
protons of 18 as those at δH 7.42 (t), 7.33 (d), 7.00 (t) and 6.79
(d) and this assignment was confirmed by a COSY experiment.
The imine proton at δH 8.08 was related via a NOESY experi-
ment to the benzenoid doublet proton at δH 7.33, which was
itself adjacent to the triplet at δH 7.00. Other NOESY corre-
lations allowed the assignments shown in Fig. 1a which were
1
extended into the 13C dimension by a H–13C HMQC experi-
ment whose results are shown in Fig. 1b. The 13C NMR
spectrum also showed the presence of a quaternary signal at
δC 92.24 comparable with the data for C2 of the known 1,3-
benzoxazines 19–21 (δC 94.10–94.33).11 Final confirmation of
the structure 18 was derived from an HMBC experiment
tailored to show longer-range proton–carbon correlations than
those revealed by the HMQC. The key correlations (Fig. 1c)
showed that the imine proton at δH 8.08 was coupled not only to
carbon atoms in the benzene ring, but also, crucially, to the
quaternary carbon at δC 92.24. This quaternary signal also
correlated with the methyl protons at δH 1.53. The linking
of the substructures shown in Fig. 1c was therefore established.
Finally, after this sequence of NMR experiments had been
carried out, it was observed that the intensity of the methyl
singlet in the 1H NMR spectrum had diminished relative to the
In contrast, the 1H NMR spectrum of the pyrolysate
obtained by FVP of the isopropoxy compound 10 at 650 ЊC was
quite different in character. Only a small aldehyde peak was
present which was assumed to be due to 4a (ca. 3%) by analogy
with the previous results. Instead, significant singlets at δH 8.08
and 1.53 (ratio 1 : 6) were observed; this product was apparently
unstable in chloroform solution. In addition, it could not be
isolated by dry-flash chromatography under our standard con-
Scheme 3 Reagents and conditions: (i) FVP (650 ЊC, ca. 0.01 Torr).
J. Chem. Soc., Perkin Trans. 1, 2001, 2704–2710
2705

View Article Online
Scheme 5 Reagents and conditions: (i) bromobenzene, 156 ЊC.
Fig. 1 NMR spectroscopic parameters of the 1,3-benzoxazine 18.
13C NMR spectrum of 22 showed aliphatic (δC 91.35) and
iminic (δC 162.36) quaternary signals perfectly compatible with
the data reported for 18 and 19–21. Moreover, a signal at
δH 1.66 is conveniently explained by the C-dimethyl group and
the base peak of the mass spectrum was formed by a loss of
15 Da from the molecular ion, as found for 18.
other resonances and that a new singlet had appeared adjacent
to the deuteriated acetone solvent multiplet. This behaviour can
be readily explained by exchange with solvent of the ketone
sub-unit of the aminal moiety, and the result lends further
support to the proposed structure (Scheme 4).
Discussion
For synthetic purposes, the most useful results from this study
are the novel gas- and solution-phase cyclisation reactions to
form the benzoxazines 18 and 22. As mentioned above, the
most likely mechanism of their formation involves a 1,6-
hydrogen atom shift, followed by cyclisation and consolidation
by loss of a hydrogen atom (Scheme 6). The 1,6-hydrogen atom
Scheme 4
Scheme 6
In order to probe this heterocyclisation reaction further, the
iminyl 1e was generated in solution, using conditions under
which it was hoped that both the degradation to 2-cyanophenol
and fragmentation to nitrile 17 should not compete. When
perester 13 was decomposed in boiling bromobenzene, a
mixture of four products was obtained, consisting of one major
component (see below) together with the ketone 7b (12%), the
oxime 11 (6%) and the oxime ether 23 (13%) as minor com-
ponents (Scheme 5). The structure of compound 23, expected
on the basis of our previous results,1,4 was clearly confirmed by
its 1H NMR spectrum, which showed the characteristic signals
of the isopropyl group (a doublet and a septuplet at δH 1.05 and
4.41, respectively), the tert-butyl group (a singlet at δH 1.20),
and, above all, the O–CH2–O moiety (a singlet at δH 5.34).
The main reaction product, obtained in an unusually high
yield (40%) for this kind of reaction, was the 1,3-benzoxazine
22. This was by far the main product both before and after
chromatographic separation, as proved by GC-MS analysis of
the crude reaction mixture. Unlike the 4-unsubstituted ana-
logue 18, compound 22 was stable during chromatography and
also when kept in chloroform solution for many hours. The
structure of 22 was established by the similarity of its spectro-
scopic characteristics with those of 18. In particular, the
shift, although still rather uncommon, has been observed12
under a variety of conditions, including FVP experiments,3,12k
and with many sorts of carbon- or heteroatom-centred radicals.
Even theoretical studies have been carried out,12f,g,j showing that
1,6-H shifts have activation barriers comparable12f or only
slightly higher12g than those of the more usually encountered
1,5-H translocations. Literature references include carbon-to-
carbon,12a–g carbon-to-oxygen,12h–k oxygen-to-carbon,12l carbon-
to-sulfur,12m and carbon-to-nitrogen3,12n,o migrations. In partic-
ular, only two examples have been reported to date concerning
the intermediacy of nitrogen-centred radicals: one deals with
the aminium radical ions involved in the Hofmann–Löffler
reaction12n and the other with the sulfonamidyl radicals pro-
duced during the oxidative chlorination of alkanesulfonamides
with the sodium persulfate–copper() chloride system.12o To
our knowledge, our reactions—including that previously
reported3—are the first examples of 1,6-hydrogen atom shifts
towards an iminyl radical. Although we have not explored
the synthetic potential of iminyl cyclisations as a route to the
1,3-benzoxazine system, compounds 18 and 22 are, respectively,
the first 4-unsubstituted benzoxazine containing only alkyl
substituents and the first 2,2-disubstituted 4-aryl derivative.
2706
J. Chem. Soc., Perkin Trans. 1, 2001, 2704–2710

View Article Online
The absence of the corresponding benzoxazines in the gas-
phase pyrolyses of the methoxy and ethoxy compounds 8 and 9
may be due to rapid hydrolysis of the ring system in solution. In
agreement with this interpretation, the level of salicylaldehyde
4a, the ultimate hydrolysis product of the benzoxazines, is
particularly high in these two reactions.
high resolution mass spectra—were not performed, the purity
of the compounds was confirmed by the absence of any
1
significant extraneous peak in the H NMR spectra and/or by
GC-MS analysis.
2-Methoxybenzaldehyde O-methyloxime 8
The other major products of the gas-phase reactions are the
nitriles 14 and 15–17. It is well known13 that iminyls can give
rise to nitriles by an α-cleavage mechanism and so the forma-
tion of the low levels of alkoxynitriles 15–17 is not unexpected.
However, the large amounts of 2-cyanophenol 14 as a major
product in all three pyrolyses cannot be explained in this way.
In particular, control experiments showed that 2-methoxy-
benzonitrile 15 did not undergo dealkylation to 2-cyanophenol
14 under the conditions of the original pyrolysis. As one
possible mechanism, we suggest that the iminyls 1b–d can
undergo a 1,5-alkyl shift under the FVP conditions to generate
the phenoxyls 24. We have previously observed2 that such
phenoxyls with an adjacent aldimine function can provide
nitriles by rearrangement to imidoyls 25 followed by α-cleavage
(Scheme 7). Alternatively, hydrogen capture—a well known
A solution of 2-methoxybenzaldehyde 5 (1.13 g, 8.3 mmol) and
O-methylhydroxylamine hydrochloride (1.40 g, 16.8 mmol) in
ethanol (35 cm3) containing pyridine (1.34 g, 14.7 mmol) was
heated under reflux for 1 h. The reaction mixture was allowed to
cool and concentrated under reduced pressure. The residue was
treated with dilute hydrochloric acid [HCl (5 drops) in water
(150 cm3)] and extracted with ether (3 × 70 cm3) to remove
remaining traces of pyridine. The organic extracts were com-
bined, dried (MgSO4) and concentrated to give the crude oxime
ether as a yellow oil. Distillation gave a single isomer of 8 as a
clear liquid (1.11 g, 81%), bp 72–76 ЊC (1.5 Torr) [lit.,14 125–127
ЊC (16 Torr)], δH 8.45 (1H, s), 7.78 (1H, dd, J 7.7 and J 0.7),
7.33 (1H, m), 6.91–6.86 (2H, m), 3.96 (3H, s) and 3.83 (3H, s);
δC 157.38 (quat.), 144.67, 130.94, 126.23, 120.61, 110.88, 61.74
(CH3) and 55.39 (CH3) (one quaternary signal overlapping)
(spectra consistent with literature data15); m/z 165 (Mϩ, 33%),
119 (100), 91 (85) and 77 (54).
3
4
2-Ethoxybenzaldehyde 6
Salicylaldehyde 4a (7.52 g, 59.7 mmol) was added to a sus-
pension of potassium carbonate (16.52 g, 120 mmol) in DMF
(100 cm3). Bromoethane (6.71 g, 61 mmol) was then added
dropwise and the solution was stirred at room temperature for
21 h. Water (100 cm3) was added and the mixture was extracted
with ether (3 × 100 cm3). The combined organic extracts were
dried (MgSO4) and concentrated under reduced pressure to
give the crude aldehyde. This was distilled to give pure 6 as a
clear liquid (7.89 g, 88%), bp 58–60 ЊC (0.5 Torr) [lit.,8 137 ЊC
(24 Torr)], δH 10.48 (1H, s), 7.80 (1H, m), 7.49 (1H, m), 7.01–
6.92 (2H, m), 4.16–4.07 (2H, q, 3J 7.0) and 1.45 (3H, t, 3J 7.0);
δC 189.80, 161.19 (quat.), 135.74, 128.02, 124.61 (quat.), 120.29,
112.29, 63.93 (CH2) and 14.49 (CH3) (spectra consistent
with literature data8); m/z 150 (Mϩ, 44%), 121 (100), 104 (16)
and 76 (12).
Scheme 7
reaction of phenoxyls2—leads to the imines 26 which may pro-
vide an alternative source of salicylaldehyde 4a by hydrolysis
(Scheme 7).
Conclusions
In conclusion, the work described in this paper has demon-
strated for the first time that iminyl radicals can interact with
ortho-alkoxy substituents under both gas- and solution-phase
conditions. The cyclisation to the benzoxazines 18 and 22 has
established the principle of a new route to these unusual hetero-
cycles, particularly under solution-phase conditions where
other reactions of the iminyl do not interfere. In the gas-phase,
these ‘other reactions’ include a surprisingly mild dealkylation
process, which may take place by an unusual alkyl group
transfer mechanism.
2-Ethoxybenzaldehyde O-methyloxime 9
A solution of 2-ethoxybenzaldehyde 6 (1.53 g, 10.2 mmol) and
O-methylhydroxylamine hydrochloride (1.74 g, 20.7 mmol)
in ethanol (50 cm3) containing pyridine (1.64 g, 20.7 mmol) was
heated under reflux for 1 h. Following the standard work up
(described for 8 above) the crude oxime ether was distilled to
give a single isomer of 2-ethoxybenzaldehyde O-methyloxime 9
as a clear liquid (1.60 g, 87%), bp 86–88 ЊC (0.8 Torr) (Found:
Mϩ, 179.0947. C10H13NO requires M, 179.0946); δH 8.53
(1H, s), 7.84 (1H, dd, 3J 7.8 and 4J 1.7), 7.35 (1H, m), 6.99–6.89
3
(2H, m), 4.11–4.06 (2H, q, J 7.0), 4.01 (3H, s) and 1.45 (3H,
t, 3J 7.0); δC 157.41 (quat.), 145.35, 131.44, 126.71, 121.28
(quat.), 121.06, 112.48, 64.41 (CH2), 62.25 (CH3) and 15.17
(CH3); m/z 179 (Mϩ, 13%), 150 (32), 133 (40), 121 (100), 91 (21)
and 77 (14).
Experimental
1H and 13C NMR spectra are recorded at 250 (or 200) and 63
(or 50) MHz respectively for solutions in [2H]chloroform
unless otherwise stated. Coupling constants are quoted in Hz;
13C NMR signals refer to CH resonances unless otherwise
stated. The 1H–13C HMQC and HMBC experiments were per-
formed using a Bruker DPX 360 NMR spectrometer, operating
at 360.13 MHz for 1H acquisitions, using the INV4GP
and INV4GPLPLRND sequences respectively, as provided by
Bruker Spectrospin. Mass spectra were obtained under electron
impact conditions unless otherwise stated. IR spectra were
recorded in chloroform solution on a Perkin-Elmer Spectrum
RX I FT-IR spectrophotometer. Column chromatography was
carried out on silica (H) using a hexane–ethyl acetate gradient
as eluent unless otherwise stated. When elemental analyses—or
2-Isopropoxybenzaldehyde 7a
Salicylaldehyde 4a (7.54 g, 61.8 mmol) was added to a sus-
pension of potassium carbonate (17.01 g, 123 mmol) in DMF
(100 cm3). 2-Bromopropane (7.52 g, 61 mmol) was then added
dropwise and the solution was stirred at room temperature for
21 h. Water (100 cm3) was added and the mixture was extracted
with ether (3 × 100 cm3). The combined organic extracts
were washed with dilute sodium hydroxide solution (2 M, 3 ×
50 cm3), to remove unreacted salicylaldehyde, and the organic
layer was dried (MgSO4) and concentrated to give the crude
J. Chem. Soc., Perkin Trans. 1, 2001, 2704–2710
2707

View Article Online
aldehyde. This was distilled to give pure 7a as a clear liquid
(2.66 g, 27%), bp 52–54 ЊC (0.4 Torr) [lit.,16 83 ЊC (1.8 Torr)];
δH 10.47 (1H, s), 7.80 (1H, dd, 3J 7.9 and 4J 1.8), 7.50 (1H, m),
2-Methoxybenzonitrile 15—control pyrolyses
FVP of 2-methoxybenzonitrile 15 (75 mg, 0.56 mmol),
80–90 ЊC, 650 ЊC, 10Ϫ2 Torr, 15 min: the NMR spectrum of the
pyrolysate showed complete recovery of starting material.
A repeat pyrolysis at a furnace temperature of 750 ЊC showed
ca. 10% decomposition and at 850 ЊC ca. 70% decomposition
was observed.
3
6.99–6.93 (2H, m), 4.66 (1H, sept, J 6.1) and 1.38 (6H, d,
3J 6.1); δC 190.05, 160.44 (quat.), 135.62, 128.09, 125.50 (quat.),
120.21, 113.81, 70.59 and 21.81 (CH3) (spectra consistent with
literature data16); m/z 164 (Mϩ, 38%), 121 (100), 104 (24),
93 (13) and 74 (33).
Pyrolysis of 2-ethoxybenzaldehyde O-methyloxime 9
2-Isopropoxybenzaldehyde O-methyloxime 10
FVP of 2-ethoxybenzaldehyde O-methyloxime 9 (0.169 g,
0.9 mmol), 50–80 ЊC, 650 ЊC, 3 × 10Ϫ2 Torr, 20 min (all products
identified from 13C NMR spectra of crude pyrolysate; absolute
yields estimated by comparison with previous pyrolyses): 2-
cyanophenol 14 (ca. 45%); δC 159.73 (quat.), 134.29, 132.87,
119.79, 117.04 (quat.), 116.59 and 99.74; salicylaldehyde 4a (ca.
15%); δC 196.54, 136.91, 134.29, 120.38, 117.48 (2 quaternaries
not apparent). No further separation was carried out (spectra
consistent with those listed above). In addition, the presence of
2-ethoxybenzonitrile 16 (ca. 10%) was inferred from ethoxy
group signals at δH 1.47 and 4.14, and at δC 14.40 and 64.49 in
the spectrum of the crude pyrolysate, which correspond closely
to those of literature data.8
A solution of 2-isopropoxybenzaldehyde 7a (1.62 g, 9.8 mmol)
and O-methylhydroxylamine hydrochloride (1.67 g, 19.9 mmol)
in ethanol (50 cm3) containing pyridine (1.60 g, 20.2 mmol)
was heated under reflux for 1 h. Following the standard work
up (described for 8 above) the crude oxime ether was distilled
to give a single isomer of 2-isopropoxybenzaldehyde O-methyl-
oxime 10 as a clear liquid (1.70 g, 90%), bp 90–92 ЊC (0.9 Torr)
(Found: Mϩ, 193.1101. C11H15NO requires M, 193.1102);
δH 8.47 (1H, s), 7.79 (1H, dd, J 7.7 and J 1.8), 7.28 (1H, m),
6.94–6.86 (2H, m), 4.55 (1H, sept, 3J 6.2), 3.96 (3H, s) and 1.32
(6H, d, 3J 6.2); δC 155.84 (quat.), 144.98, 130.79, 126.23, 121.61
(quat.), 120.47, 113.69, 70.66 (CH), 61.69 (O-CH3) and 21.91
(CH3); m/z 193 (Mϩ, 23%), 151 (41), 119 (100) and 91 (72).
3
4
Pyrolysis of 2-isopropoxybenzaldehyde O-methyloxime 10
2-Isopropoxybenzonitrile 17
FVP of 2-isopropoxybenzaldehyde O-methyloxime 10 (0.450 g,
2.3 mmol), 60–90 ЊC, 650 ЊC, 3 × 10Ϫ2 Torr, 30 min: 2-cyano-
phenol 14 (0.155 g, 34%), mp 84–86 ЊC, mixed mp 91 ЊC (lit.,8
94 ЊC); δH 7.48–7.41 (2H, m) and 6.98–6.92 (2H, m) (OH signal
not apparent); δC 160.33 (quat.), 134.39, 132.84, 119.40, 117.22
(quat.), 116.68 and 99.30 (quat.) (spectra consistent with litera-
ture data8); m/z 119 (Mϩ, 58%), 91 (100), 64 (35) and 38 (21):
2,2-dimethyl-2H-1,3-benzoxazine 18 (ca. 0.125 g, 34%) (1H and
13C NMR spectra obtained as a 1 : 1 mixture with 14 at 360
and 90 MHz respectively in [2H6]acetone); δH 8.08 (1H, s), 7.42
(1H, ddd, J1 7.4, J2 8.2 and J 1.7), 7.33 (1H, dd, J 7.3 and
4J 1.7), 7.00 (1H, overlapping m), 6.79 (1H, d, 3J 8.2) and 1.53
(6H, s); δC 152.70, 132.80, 126.37, 120.12, 115.30, 92.24 (quat.)
and 25.98 (CH3) (2 aromatic quaternaries not assigned);
m/z (GC-MS) 161 (Mϩ, 46%), 146 (100), 120 (43), 91 (23) and
77 (22); 2-isopropoxybenzonitrile 17 (0.041 g, 10%) δH 7.57–
7.45 (2H, m), 6.98–6.91 (2H, m), 4.64 (1H, m) and 1.39 (6H, d,
3J 6.1); δC 159.80 (quat.), 133.97, 133.82, 120.32, 116.70 (quat.),
113.53, 102.84 (quat.), 71.67 and 21.72 (CH3); m/z (GC-MS)
161 (Mϩ, 9%), 119 (100), 91 (31) and 64 (8) (spectra compatible
with those of the authentic sample reported above); a trace of
salicylaldehyde 4a (ca. 3%) was identified by the presence of its
aldehyde signal (δH 9.88) in the 1H NMR spectrum of the crude
pyrolysate.
2-Cyanophenol (200 mg, 1.7 mmol), followed by 2-bromo-
propane (207 mg, 1.7 mmol), was added to a suspension of
potassium carbonate (340 mg, 3.4 mmol) in DMF (3 cm3) and
the solution was stirred overnight. Water (6 cm3) was added
and the mixture was extracted with ether (3 × 6 cm3). The com-
bined organic extracts were washed with water (6 cm3) and the
organic layer was dried (MgSO4) and concentrated to give the
crude nitrile 17, bp 146–149 ЊC (15 Torr) [lit.,10 137–138 ЊC (13
Torr)] (Found: Mϩ 161.0841. C10H11NO requires M 161.0841);
δH 7.53–7.44 (2H, m), 6.97–6.90 (2H, m), 4.63 (1H, sept, 3J 6.1)
3
3
4
3
3
and 1.38 (6H, d, J 6.1); δC 159.74 (quat.), 133.97, 133.73,
120.28, 116.60 (quat.), 113.48, 102.75 (quat.), 71.59 and 21.66
(CH3); m/z 161 (Mϩ, 9%), 119 (100), 91 (47), 64 (18) and 63 (17).
Pyrolysis experiments
Precursors were sublimed or distilled under vacuum through
the furnace tube which was held at the stated temperature.
Conditions are quoted in the following manner: quantity of
substrate, inlet temperature, furnace temperature, pressure,
pyrolysis time, products (most abundant first). Products were
trapped in liquid nitrogen. The work up of the pyrolyses
involved the dissolution of the pyrolysate in dichloromethane
and then separation by dry flash chromatography over silica
using hexane–ethyl acetate mixtures as eluent.
(2-Isopropoxyphenyl)(phenyl)methanone oxime 11
According to a standard procedure,17 a solution of (2-isoprop-
oxyphenyl)(phenyl)methanone (7b)18 (7.68 g, 32 mmol) and
hydroxylamine hydrochloride (6.62 g, 96 mmol) in a mixture of
ethanol (80 cm3) and pyridine (8 cm3) was heated under reflux
overnight under mechanical stirring. Most of the solvent was
evaporated and the resulting mixture was poured into ice–water
and extracted with diethyl ether. The organic phase was dried,
the solvent was removed and the residue chromatographed
(light petroleum–diethyl ether gradients) to give the title com-
pound 11 (7.34 g, 90%) as a 4 : 1 mixture of geometric isomers,
mp 108–110 ЊC (lit.,19 111 ЊC); δH 9.06 (1H, br s, OH, both
isomers), 7.60–7.20 (m, Ar-H, both isomers), 7.04 (1H, ddd,
Pyrolysis of 2-methoxybenzaldehyde O-methyloxime 8
FVP of 2-methoxybenzaldehyde O-methyloxime 8 (0.365 g,
2.2 mmol), 30–70 ЊC, 650 ЊC, 3 × 10Ϫ2 Torr, 20 min: dry-flash
chromatography gave 2-cyanophenol 14 (0.115 g, 46%), mp
82–83 ЊC, mixed mp 89–91 ЊC (lit.,8 94 ЊC); δH 7.50–7.41 (2H,
m), 7.00–6.92 (2H, m) and 5.62 (1H, br s); δC 158.72 (quat.),
134.66, 132.85, 120.65, 116.50 and 99.21 (one quaternary not
apparent) (spectra consistent with literature data8); m/z 119
(Mϩ, 75%), 91 (85), 64 (100) and 39 (7). In addition, salicylalde-
hyde 4a (0.062 g, 24%), δH 9.89, was obtained, whose presence
was unambiguously confirmed from the 13C NMR spectrum
of crude pyrolysate, δC 196.54, 161.57 (quat.), 136.91, 133.64,
120.63 (quat.), 119.75 and 117.49 (spectra consistent with litera-
ture data8): the presence of a trace of 2-methoxybenzonitrile 15
(ca. 5%) was inferred from the methoxy group signal at δH 3.92
in the spectrum of the crude pyrolysate [δH (authentic sample)
3.92].
3
4
3J1, J2 7.3 and J3 1.0, Ar-H, major isomer), 6.99 (1H, br d,
3
3
3J 8.7, Ar-H, major isomer), 6.93 (1H, ddd, J1, J2 7.3 and
3
4J3 1.2, Ar-H, minor isomer), 6.82 (1H, br d, J 8.0, Ar-H,
3
minor isomer), 4.45 (1H, sept, J 6.2, CH, major isomer), 4.33
3
3
(1H, sept, J 5.9, CH, minor isomer), 1.10 (6H, d, J 6.2, Me,
major isomer), and 0.92 (6H, d, 3J 5.9, Me, minor isomer).
2708
J. Chem. Soc., Perkin Trans. 1, 2001, 2704–2710

View Article Online
2-{[(2-Isopropoxyphenyl)(phenyl)methylidene]aminooxy}acetic
acid 12
the residue chromatographed (light petroleum–diethyl ether
gradients) to give, in order of elution, 2,2-dimethyl-4-phenyl-
2H-1,3-benzoxazine 22 (1.38 g, 40%), as a thick oil that slowly
crystallises in square prisms, mp 82–83 ЊC (from light petrol-
eum) (Found: C, 90.3; H, 6.3; N, 5.8. C16H15NO requires C,
90.0; H, 6.4; N, 5.9%. Found: Mϩ 237.1162. C16H15NO requires
M, 237.1154); νmax/cmϪ1 2985 and 1614; δH (300 MHz) 7.57–
7.53 (2H, m, Ar-H), 7.47–7.42 (3H, m, Ar-H), 7.36 (1H, ddd,
According to the reported procedure,20 a mixture of oxime 11
(7.15 g, 28 mmol), chloroacetic acid (4.25 g, 45 mmol) and
sodium hydroxide (3.36 g, 84 mmol) in water (25 cm3)–ethanol
(15 cm3) was heated under reflux overnight. Then it was poured
into ice–water and neutralised with conc. hydrochloric acid.
The mixture was extracted with dichloromethane, the organic
phase was dried, the solvent removed and the residue chroma-
tographed (light petroleum–diethyl ether gradients) to give the
title compound 12 (5.82 g, 66%) as a practically pure single iso-
mer, mp 137–138 ЊC (from light petroleum–benzene) (Found:
C, 69.3; H, 6.0; N, 4.45. C18H19NO4 requires C, 69.0; H, 6.1; N,
4.5%. Found: Mϩ 313.1318. C18H19NO4 requires M, 313.1314);
νmax/cmϪ1 3286, 1769, 1487, 1447, 1349 and 1325; δH 9.70 (1H,
br s, OH), 7.55–7.27 (6H, m, Ar-H), 7.18–7.09 (3H, m, Ar-H),
4.74 (2H, s, CH2), 4.56 (1H, sept, 3J 6.0, CH), and 1.22 (6H, d,
3J 6.0, Me); δC 172.48 (quat.), 155.18 (quat.), 135.20 (quat.),
131.30 (quat.), 130.72, 129.86, 129.01, 128.13, 128.00, 125.03
(quat.), 122.72, 117.23, 75.51 (CH), 71.98 (CH2), and 22.75
(CH3); m/z 313 (Mϩ, 12%), 271 (12), 238 (33), 196 (66), 195
(100), 167 (54) and 77 (47).
3
4
3
3J1 9.0, J2 7.3 and J3 1.7, Ar-H), 7.18 (1H, dd, J1 7.5 and
4J2 1.5, Ar-H), 6.95–6.80 (2H, m, Ar-H), and 1.66 (6H, s, Me);
δC (75 MHz) 162.36 (quat.), 155.77 (quat.), 137.80 (quat.),
134.11, 130.21, 129.37, 128.97, 128.46, 121.22, 117.72, 117.56
(quat.), 91.35 (quat.) and 27.61 (CH3); m/z 237 (Mϩ, 16%),
236 (27), 222 (100), 153 (15), 152 (34), 77 (28) and 76 (39);
(2-isopropoxyphenyl)(phenyl)methanone
O-(tert-butoxy-
methyl)oxime 23 (0.62 g, 13%) as a single geometric isomer,
oil (Found: C, 74.1; H, 8.2; N, 4.15. C21H27NO3 requires C, 73.9;
H, 8.0; N, 4.1%); νmax/cmϪ1 2976, 1598, 1486, 1450, 1243, 1119,
1012 and 961; δH (300 MHz) 7.54–7.47 (2H, m, Ar-H), 7.36–
3
4
7.21 (4H, m, Ar-H), 7.19 (1H, dd, J1 7.5 and J2 1.8, Ar-H),
6.98 (1H, dd, 3J1 7.4 and 4J2 0.8, Ar-H), 6.93 (1H, br d, 3J 8.5,
Ar-H), 5.34 (2H, s, CH2), 4.41 (1H, sept, 3J 6.0, CHMe2), 1.20
(9H, s, tert-Bu), and 1.05 (6H, d, 3J 6.0, CHMe2); δC (75 MHz)
156.42 (quat.), 155.34 (quat.), 136.94 (quat.), 130.86, 130.28,
129.45, 128.50, 127.70, 125.05 (quat.), 120.59, 114.47, 93.38
(CH2), 75.32 (quat.), 71.03, 29.35 (CH3), and 22.47 (CH3);
m/z 315 (Mϩ Ϫ 86, 1%), 238 (29), 223 (34), 196 (43), 195 (66),
167 (33), 77 (62) and 57 (100); ketone 7b (0.42 g, 12%); oxime 11
(0.22 g, 6%).
tert-Butyl 2-{[(2-isopropoxyphenyl)(phenyl)methylidene]amino-
oxy}peracetate 13
According to the reported procedure,21 the iminoxyacetic acid
12 (5.82 g, 18.6 mmol) was added at room temperature and
under nitrogen to a stirred solution of CDI (3.01 g, 18.6 mmol)
in anhydrous THF (70 cm3). After 1 h, a solution of tert-butyl
hydroperoxide (4.31 g, 33.5 mmol) in light petroleum (30 cm3)
was added dropwise at 0 ЊC and the mixture was kept at 0–5 ЊC
for 4 h. tert-Butyl hydroperoxide (Aldrich) was used as a 70
wt% solution in water, previously dried by extraction with cold
light petroleum. The mixture was poured into ice–water and
extracted with cold diethyl ether. The organic phase was washed
twice with cold water and dried. After removal of the solvent
the residue was chromatographed (light petroleum–diethyl
ether gradients) to give the title perester 13 (5.62 g, 78%) as
a thick oil which was a 5 : 1 mixture of geometric isomers;22
νmax/cmϪ1 3000, 1792, 1598, 1486, 1451, 1368, 1246 and 1089;
δH (300 MHz) 7.52–7.24 (m, Ar-H, both isomers), 7.02 (1H,
Acknowledgements
This investigation was supported by the University of Bologna
(1999–2001 Funds for Selected Research Topics) and by the
Italian Ministry of Scientific and Technological Research
(MURST, 2000–2001 Grant for “Radical Processes in
Chemistry and Biology: Syntheses, Mechanisms, and Applica-
tions”). We thank the EPSRC, Bruker Spectrospin and Glaxo
Wellcome for support for the DPX 360 NMR spectrometer.
References
3
3
4
ddd, J1, J2 7.4 and J3 1.0, Ar-H, major isomer), 6.98–6.92
1 G. Calestani, R. Leardini, H. McNab, D. Nanni and G. Zanardi,
J. Chem. Soc., Perkin Trans. 1, 1998, 1813.
2 M. Black, J. I. G. Cadogan, R. Leardini, H. McNab, G. McDougald,
D. Nanni, D. Reed and A. Zompatori, J. Chem. Soc., Perkin Trans.
1, 1998, 1825.
3 R. Leardini, H. McNab, D. Nanni, S. Parsons, D. Reed and
A. G. Tenan, J. Chem. Soc., Perkin Trans. 1, 1998, 1833.
4 R. Leardini, H. McNab, M. Minozzi and D. Nanni, J. Chem. Soc.,
Perkin Trans. 1, 2001, 1072.
5 T. Creed, R. Leardini, H. McNab, D. Nanni, I. S. Nicolson and
D. Reed, J. Chem. Soc., Perkin Trans. 1, 2001, 1079.
6 J. I. G. Cadogan, C. L. Hickson and H. McNab, Tetrahedron, 1986,
42, 2135.
3
(1H ϩ 1H, br d, J 8.2, Ar-H, major isomer, overlapped with
ddd, 3J1, 3J2 7.5 and 4J3 1.0, Ar-H, minor isomer), 6.81 (1H, br
d, 3J 8.2, Ar-H, minor isomer), 4.81 (2H, s, CH2, minor isomer),
4.76 (2H, s, CH2, major isomer), 4.43 (1H, sept, 3J 6.0, CHMe2,
3
major isomer), 4.33 (1H, sept, J 6.0, CHMe2, minor isomer),
1.32 (9H, s, tert-Bu, both isomers), 1.06 (6H, d, 3J 6.0, CHMe2,
major isomer), and 0.88 (6H, d, 3J 6.0, CHMe2, minor isomer);
m/z 385 (Mϩ, <1%), 341 (4), 311 (3), 255 (6), 238 (90), 237 (49),
236 (39), 223 (100), 222 (95), 196 (87), 195 (97), 181 (35), 167
(40), 152 (36), 77 (63) and 57 (98).
CAUTION: since hydroperoxides and peresters are poten-
tially hazardous compounds, they must be handled with due
care; avoid exposure to strong heat or light, mechanical shock,
oxidizable organic materials, or transition metal ions. No
particular difficulties were experienced in handling of the new
perester synthesised in this work using the procedure described
above. Even its column chromatography did not give any
problem; however, we advise that the separation should be
carried out with extreme care, evaporating the solvent under
reduced pressure with a water bath kept below 20 ЊC.
7 The peracetate 13 was prepared according to Forrester’s method (see
refs. 1 and 4 and references therein), but the final esterification step
was carried out by direct treatment of the acid with tert-butyl
hydroperoxide and N,NЈ-carbonyldiimidazole without previous
conversion of the acid into an acyl halide.
8 J. Pouchert and J. Behnke, The Aldrich Library of 1H and 13C
FTNMR Spectra, Edition 1, Aldrich Chemical Company Inc.,
1993.
9 R. A. Aitken, G. Burns and J. J. Morrison, J. Chem. Soc., Perkin
Trans. 1, 1998, 3937 and references therein.
10 O. L. Mndzhoyan and G. M. Pogosyan, Izv. Akad. Nauk Arm. SSR,
Khim. Nauki, 1963, 16, 264 (Chem. Abstr., 1964, 60, 6780).
11 H. Gröger, M. Hatam and J. Martens, Tetrahedron, 1995, 51,
7173.
Decomposition of perester 13
12 (a) J. Y. Nedelec and D. Lefort, Tetrahedron Lett., 1972, 5073;
(b) B. C. Gilbert and D. J. Parry, J. Chem. Soc., Perkin Trans. 2, 1986,
1345; (c) J. A. Franz, D. H. Roberts and K. F. Ferris, J. Org. Chem.,
1987, 52, 2256; (d ) R. K. Freidlina, N. V. Blinova, V. I. Dostovalova
and E. T. Chukovskaya, Zh. Obshch. Khim., 1989, 59, 851 (Chem.
A solution of the perester 13 (5.62 g, 14.6 mmol) in bromo-
benzene (50 cm3) was added dropwise over a period of 1 h to
boiling bromobenzene (700 cm3). After one additional hour at
reflux, the solution was cooled, the solvent was evaporated and
J. Chem. Soc., Perkin Trans. 1, 2001, 2704–2710
2709

View Article Online
14 J. C. Irvine and A. M. Moodie, J. Chem. Soc., 1908, 93, 102.
15 C.-J. Chang, T.-L. Shieh and H. G. Floss, J. Med. Chem., 1977, 20,
176.
16 T. Horaguchi, C. Tsukada, E. Hasegawa, T. Shimizo, T. Suzuki and
K. Tanemura, J. Heterocycl. Chem., 1991, 28, 1261.
17 Vogel’s Textbook of Practical Organic Chemistry, 4th edn.,
Longman, Harlow, 1988, p. 1150.
18 H. Tomioka, K. Nakanishi and Y. Izawa, J. Chem. Soc., Perkin
Trans. 1, 1991, 465.
19 Y. Bonnard and J. Meyer-Oulif, Bull. Soc. Chim. Fr., 1931, 1303.
20 A. R. Forrester, M. Gill, C. J. Meyer, J. S. Sadd and R. H. Thomson,
J. Chem. Soc., Perkin Trans. 1, 1979, 606.
21 L. A. Singer and N. P. Kong, J. Am. Chem. Soc., 1967, 89, 5251;
H. A. Staab, Angew. Chem., Int. Ed. Engl., 1962, 1, 351.
22 Due to the very low molecular ion and decomposition hazard,
neither a high resolution mass spectrum nor an elemental analysis
was obtained for this perester; its purity was however confirmed
by the complete absence of any significant extraneous peak in its 1H
NMR spectrum.
Abstr., 1990, 112, 7562); (e) N. V. Blinova and R. G. Gasanov,
Kinet. Katal., 1990, 31, 524 (Chem. Abstr., 1990, 113, 212079);
( f ) X. L. Huang and J. J. Dannenberg, J. Org. Chem., 1991, 56, 5421;
(g) B. Viskolcz, G. Lendvay and L. Seres, J. Phys. Chem. A, 1997,
101, 7119; (h) C. G. Francisco, R. Freire, R. Hernandez,
M. C. Medina and E. Suarez, Tetrahedron Lett., 1983, 24, 4621;
(i) G. I. Nikishin, I. V. Svitanko and E. I. Troyansky, J. Chem. Soc.,
Perkin Trans. 2, 1983, 595; (j) T. P. W. Jungkamp, J. N. Smith and
J. H. Seinfeld, J. Phys. Chem. A, 1997, 101, 4392; (k) P. F. Britt,
A. C. Buchanan, III, M. J. Cooney and D. R. Martineau,
J. Org. Chem., 2000, 65, 1376; (l ) P. M. Blum, A. G. Davies and
R. A. Henderson, J. Chem. Soc., Chem. Commun., 1978, 569; (m) M.
Odabasoglu, R. Tapramaz and I. E. Gumrukcuoglu, Phosphorus,
Sulfur Silicon Relat. Elem., 2000, 158, 57; (n) E. J. Corey and
W. R. Hertler, J. Am. Chem. Soc., 1960, 82, 1657; (o) G. I. Nikishin,
E. I. Troyansky and M. I. Lazareva, Tetrahedron Lett., 1985, 26,
3743.
13 K. Bird, A. W. K. Chan and W. D. Crow, Aust. J. Chem., 1976, 29,
2281.
2710
J. Chem. Soc., Perkin Trans. 1, 2001, 2704–2710