Ring forming reactions of imines of 2-aminobenzaldehyde and related compounds

Article abstract of DOI:10.1039/b209505j

By addition of organometallic reagents to 2-aminobenzonitriles followed by quenching with suitable electrophiles (acyl halides, aldehydes and ketones), several types of 6-membered benzofused aromatic and non-aromatic nitrogen heterocycles could be obtained. Rearrangements leading to 1,2-dihydro-3H-1,4-benzodiazepin-3-ones and preparation of various quinazolines are described.

Full text of DOI:10.1039/b209505j

View Article Online / Journal Homepage / Table of Contents for this issue
Ring forming reactions of imines of 2-aminobenzaldehyde and
related compounds
Per Wiklund and Jan Bergman*
Unit for Organic Chemistry, Department of Biosciences,
Karolinska Institute and Södertörn University College, Novum Research Park,
SE-141 57 Huddinge, SWEDEN.
E-mail: Jan.Bergman@biosci.ki.se
Received 27th September 2002, Accepted 7th November 2002
First published as an Advance Article on the web 9th December 2002
By addition of organometallic reagents to 2-aminobenzonitriles followed by quenching with suitable electrophiles
(acyl halides, aldehydes and ketones), several types of 6-membered benzofused aromatic and non-aromatic nitrogen
heterocycles could be obtained. Rearrangements leading to 1,2-dihydro-3H-1,4-benzodiazepin-3-ones and
preparation of various quinazolines are described.
These intriguing products combined with the potential
Introduction
synthetic usefulness of the difunctional structures 2 and 3, has
prompted us to investigate these systems and their reactions in
more detail.
In this paper we report a study on the reactivity and synthetic
usefulness of the dianions 2 and the corresponding acids 3
(Scheme 1). When treated with certain electrophiles, these gave a
variety of benzofused 6- and 7-membered cyclic products.
Results and discussion
As a starting point in the investigation of the nucleophilic
properties of species of type 2 and 3, the dianion 2a and the cor-
responding 2-aminobenzophenone imine 3a [2-(iminophenyl-
methyl)benzene amine] were reacted with various electrophiles.
These specific derivatives were chosen because 2-aminobenzo-
phenone imine (3a) is readily available and described in the
literature.2 In general, imines which have a hydrogen atom on
the nitrogen are very susceptible to hydrolysis to the corre-
sponding ketone and are therefore not easy to isolate.
Thorough studies on the rate of such hydrolyses have been
published.3,4
Dianions of type 2 are formed after addition of two equiv-
alents of a metalating reagent to the corresponding anthranilo-
nitrile in THF or diethyl ether solution. In some cases the imine
anion can be carefully protonated to the imine and isolated as
such (vide supra). Addition of PhMgBr to the nitrile function-
ality is preferably performed in THF, whereas addition of metal
alkyls works best in ether. Attempts to add MeMgI or MeLi to
anthranilonitriles invariably failed, i.e. after addition of an
electrophile the nitrile function was always intact. Copper () is
known to catalyse the addition of Grignard reagents to nitriles,5
but addition of CuBr did not change the outcome. As Couture
et al. have reported, other metal alkyl reagents readily add to
N-substituted anthranilonitriles.6
Scheme 1
It has previously been shown that compound 4a, a rare
example of
a 1,4-benzodiazepine-3-one (isolated as two
separate solid state conformers, the structures of which were
determined by X-ray crystallography), together with the
quinazoline 5a, can result from treatment of anthranilonitrile
(2-aminobenzonitrile) with two equivalents of a Grignard
reagent, followed by addition of 2-bromoisobutyryl bromide
(Scheme 2).1 Alternatively, the preformed amide 6a could be
treated with a Grignard reagent to give similar results.
Reactions with acylating agents
The initial reaction of the dianions 2 with acyl halides is of
course very fast because of the high intrinsic reactivity of the
reagents. Due to the difunctional nature of 2, this initial
reaction seems inevitably to be followed by a favoured ring-
closure (6-exo-trig) to the six-membered ring of the inter-
mediate anion 7 (Scheme 3). It does not matter which nitrogen
atom reacts first as both reaction paths will lead to the same
charged intermediate. It seems reasonable to assume that the
charge resides to a significant degree on the exocyclic oxygen,
rather that on the nitrogen (even/also when R1 = H), because of
the much lower basicity of the aliphatic alkoxide compared
with the anilinic amine anion. This reasoning is based on our
collected experience with these systems, and serves as a working
model to explain the outcome of the reactions. The fate of the
Scheme 2
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 6 7 – 3 7 2
367

View Article Online
Scheme 3
intermediate 7 depends on the nature of the exocyclic sub-
stituent R and the reaction conditions at large. If R1 = H,
protonation and elimination of water will lead to formation of
an aromatic quinazoline. There are several recent reports on
synthesis of various quinazolines and quinazolinones from
anthranilonitrile. Among these are 4-aminoquinazolines,7–10
3,4-dihydro-4-thioxo-2(1H )-quinazolinones
(2-hydroxy-4-
mercaptoquinazolines),11–13
2,4(1H,3H )-quinazolinediones
(2,4-dihydroxyquinazolines) and 4-bromo-2-(chloromethyl)-
quinazoline.14
Reactions with 2-haloacyl halides
The reaction of 2a with 2-bromoisobutyryl bromide has been
reported.1 This reaction, for which a full preparative procedure
appears in this paper, yields 1,2-dihydro-2,2-dimethyl-5-phenyl-
3H-1,4-benzodiazepin-3-one (4a) and a small amount of
2-(1-bromo-1-methylethyl)-4-phenylquinazoline (5a) (Scheme
2). The former is to our knowledge the only 1,4-benzodiazepine-
3-one for which a synthesis has been reported. It has the rather
unusual property of having two distinguishable and reasonably
stable solid-state ring-conformers (structural elucidation of
these were performed by X-ray crystallography).1
It was initially discovered that 6a when treated with PhMgBr
gave 4a. In this case it is apparent that a rearrangement must
be involved because the acyl function appears next to the imino
nitrogen of the product. Later it was found that the same
product arose if anthranilonitrile was first treated with PhMg-
Br, and then with the acyl bromide (Scheme 4). It is reasonable
to assume that this unusual product is not formed through two
completely different mechanisms, and that a common inter-
mediate of type 7 (Scheme 3) is involved. In the original paper a
possible rearrangement mechanism was put forth involving
formation of an aziridinone which would be nucleophilically
opened to form the product. We now propose another mech-
anism in which a six-membered ring expands to form the
seven-membered ring of the product (Scheme 4). The latter
mechanism would also explain the formation of the halo-
methylquinazoline 5a by protonation and elimination of water.
Therefore, according to the principle of OccamЈs razor, one
should choose the second proposal. Could a rearranged
product be isolated from a reaction with a dianion derived from
N-methylanthranilonitrile, it would certainly be more tangible
evidence that this assumed mechanism is correct. In that case,
intermediate formation of a three-membered ring (with a
Scheme 4
neutral nitrogen) is not possible. Methylation of anthranilo-
nitrile was performed according to the literature,15 and the
subsequent reaction did indeed give 1,2,2-trimethyl-5-phenyl-
3H-1,4-benzodiazepin-3-one (4b) (the structure was confirmed
by methylation of 4a). Hence the conclusion that ring-
expansion is part of the mechanism. Formation of an inter-
mediate oxirane can of course not be ruled out. This picture
is somewhat clouded by the fact that treatment of the N-
methylated preformed amide 6b with PhMgBr failed to give
the benzodiazepine, possibly because of attack on the amide
followed by degradation of the system. Curiously it also
appears that only tertiary haloacyl halides will give the
7-membered ring, possibly an indication that the ring-expand-
ing substitution must be preceded by ionisation (SN1). A
hypothetical intermediate oxirane would also be more stabilised
were it more heavily substituted. This reasoning is based on the
fact that 4a was isolated in high yield, whereas no 7-membered
ring could be identified in the complex reaction mixture after
reaction of 2a with 2-bromopropionyl bromide (a methyl group
less). In contrast, the gem-diethyl substituted compound 4c
could prepared in low yield, showing more generality of the
reaction with tertiary haloacyl halides (Fig. 1). The IR-spectra
of the gem-dimethyl compound 4a and that of the gem-diethyl
compound 4c are almost identical with absorption peaks at
3299/3299, 1682/1684, 1612/1624 cmϪ1 respectively.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 6 7 – 3 7 2
368

View Article Online
of strong aluminium complexes, but it was subsequently
discovered that aqueous base could dissolve these. After extrac-
tion with an organic solvent, it also proved possible to extract
the products with aqueous acid (HCl), leaving unwanted
organic matter in the organic phase. Neutralisation of the
acidic solutions afforded the products. This procedure does
not give excellent yields, but its simplicity and the fact that
it does give access to an unusual set of compounds, speaks in
its favour. After attempted recrystallisation of 9a in hot
hexane, the known18,19 fully aromatic 2-phenylquinazoline was
recovered, demonstrating the ease of oxidation (dehydrogen-
ation) of these compounds.
Synthesis of the interesting spiro compound 10a has been
published,16 but without experimental details and a very low
yield. It was discovered that for reaction of 2a with ketones
it is beneficial to treat the dianion with two equivalents of
ammonium chloride (added as a solid to the THF-solution),
protonating the amine and the imine, prior to addition of the
ketone. Using this new procedure in conjunction with the acidic
extraction used for purification of the compounds 9, it proved
possible to isolate 10a in 70% yield without need for any chroma-
tography after reaction with cyclohexanone. Similarly, reaction
with acetone afforded the gem-dimethyl compound 10b. The
ease of protonation of the compounds 10 was somewhat
surprising as 5% citric acid could be used for the extraction.
There is a published stepwise procedure from anthranilo-
nitrile to 2,2-diaryl substituted 1,2-dihydroquinazolines,20 as
well as a mass spectrometric study of these compounds.21 These
products can formally be derived from a dianion of type 2 and
an appropriate ketone. However, benzophenone was found to
be too unreactive (hindered) to be used in this context.
Fig. 1 Products from reactions with acyl halides.
Reaction of 2a with chloroacetyl chloride gave the chloro-
methylquinazoline 5b, a better yield of which was isolated after
addition of solid NH4Cl prior to addition of the acid chloride.
The dichloromethyl compound 5c could be prepared similarly.
A two-step procedure to such compounds, which is in principle
the same reaction, has been reported.16 A very clean reaction
leading to 5b is deprotonation of the imine 3a with sodium
hydride in THF, followed by treatment with chloroacetyl
chloride. Reaction of these two substrates without any added
base, or under acidic conditions (dioxane saturated with HCl),
also led to the same product, demonstrating the great stability
of the six-membered intermediate and product. The dichloro-
methyl quinazoline 5c showed a remarkable resilience to
hydrolysis and the corresponding formylquinazoline could not
be obtained.
In reactions of N1-blocked/protected
2 with haloacyl
halides, there is as we have seen no possibility for formation of
the usual 6-membered aromatic product. However, metalation
of N-methyl- or N-benzylanthranilonitrile, followed by treat-
ment with acetyl chloride, gave very complex reaction mixtures.
Reactions with aldehydes and ketones
Conclusions
In reactions of amino dianions 2 with aldehydes, 1,2-dihydro-
quinazolines are formed via a presumed initial formation of
an imine, exemplified by compounds 8a and 8b (Fig. 2). The
Addition of organometallic reagents to 2-aminobenzonitriles,
followed by capture of the dianion with electrophiles (acyl
halides, aldehydes and ketones), is a preparatively useful
route to various quinazolines. The quinazoline skeleton appears
in alkaloids22–25 and in medicinal derivatives such as prazozin
(antihypertensive α1-adrenoreceptor antagonist)26,27 and
methaqualone (sedative).28
Using variants on the synthetic scheme described in this
paper, quinazolines or 1,2-dihydroquinazolines substituted in
position 2 with alkyl or aryl groups can be obtained. Earlier
work has shown that other electrophiles can be utilized to gain
access to e.g. 2-quinazolinones.16 Because of the apparent
inability for addition of alkyl metal reagents to N-unsub-
stituted anthranilonitriles, direct access to aromatic quin-
azolines is limited to those with an aryl substituent at position
4. However, treatment of anthranilonitriles with DIBAL-H,
followed by capture of the dianion with an aldehyde, provides
access to 4-unsubstituted quinazolines after oxidation to the
aromatic system. We have also shown that addition of alkyl
metal reagents to N-substituted anthranilonitriles, followed by
reaction with aldehydes, is a feasible route to the corresponding
1,2-dihydroquinazolines. This provides a route to 4-alkyl-1,2-
dihydroquinazolines, and opens access to the aromatic systems
by deprotection of N1.
Fig. 2 Products from reactions with aldehydes and ketones.
corresponding stepwise reaction, involving isolation of the
2-iminobenzonitrile followed by addition to the nitrile, has
been described16 and debated.17 The precursor of 8b, i.e.
When tertiary haloacyl halides were added to the 2-amino-
benzophenone dianions 2, 1,2-dihydro-3H-1,4-benzodiazepin-
3-ones were formed. These compounds are unusual,29 despite
the close structural similarity with the common 1,2-dihydro-
3H-1,4-benzodiazepin-2-ones. The mechanism of formation
appears to involve ring-expansion from a 6-membered ring.
N-benzylated
anthranilonitrile,
2-[(phenylmethyl)amino]-
benzonitrile (1a), could be prepared in a fast one-pot procedure
by reductive alkylation in 95% yield.
Treatment of anthranilonitrile with two equivalents of
DIBAL-H reduces the nitrile functionality to an imine anion.
Capture with benzaldehyde led to the formation of 1,2-dihydro-
quinazolines 9. The highest yields were obtained when two
equivalents of DIBAL-H were used, the assumption being that
the first equivalent functions as base. Both less and more
DIBAL-H diminished the yield drastically. In work-up of
these reactions there were initially problems due to formation
Experimental
All starting materials and solvents (PA grade) are commercially
available and used without further purification. THF was dried
over sodium.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 6 7 – 3 7 2
369

View Article Online
NMR spectra were recorded on a Bruker DPX 300 (1H 300
MHz, 13C 75 MHz) in DMSO-D6, using the solvent peak as
reference (1H 2.50 ppm, 13C 39.51 ppm). Coupling constants (J)
are given in Hz. IR spectra were recorded on a Perkin–Elmer
1600 FT-IR on KBr-tablets. HRMS-determinations (FAB)
were performed by Einar Nilsson, University of Lund,
Sweden. Elemental analyses were performed by H. Kolbe
Mikroanalytisches Laboratorium, Mülheim an der Ruhr,
Germany. Melting points were determined on a Reichert
Kofler hot stage.
5a formed and was filtered off. Yield 0.120 g (0.37 mmol, 3.4%)
powder, mp 136 ЊC. The material could be recrystallised from
EtOH to yield white needles, mp 137 ЊC (Found C, 62.51; H,
4.57; N, 8.53. C17H15BrN2 requires C, 62.40; H, 4.62; N, 8.56%);
IR νmax: 1608, 1561, 1541, 1485, 1464, 1390, 1332, 1164, 1096,
782, 701, 622, 596, 522 cmϪ1; δH: 2.23 (6 H, s), 7.63–7.68 (3 H,
m), 7.74–7.79 (1 H, m), 7.82–7.86 (2 H, m), 8.03–8.14 (3 H, m);
δC: 32.4 (q), 66.1 (s), 120.7 (s), 126.8 (d), 128.8 (d), 128.8 (d),
130.1 (d), 130.3 (d), 134.6 (d), 135.6 (s), 150.3 (s), 166.0 (s),
168.0 (s)
2-[(Phenylmethyl)amino]benzonitrile (1a)
1,2-Dihydro-1,2,2-trimethyl-5-phenyl-3H-1,4-benzodiazepin-3-
Anthranilonitrile (5.90 g, 50.0 mmol) was dissolved in 50 mL
AcOH. Benzaldehyde (6.5 mL, 64 mmol) was added and the
reaction stirred at rt for 30 min. The reaction flask was
immersed in ice and NaBH4 (2.0 g, 52 mmol) was added in
portions. A vigorous exothermic reaction followed on each
addition. On cooling a thick precipitate formed. After addition
of 100 mL water, the product was filtered off and dried to
yield 9.89 g (47.5 mmol, 95%) white powder, mp 118 ЊC (lit.30
mp 117–118 ЊC).
one (4b)
To a 20 mL THF-solution of 5.0 mmol 2b (from N-methyl-
anthranilonitrile and PhMgBr according to the general pro-
cedure) cooled to 0 ЊC, 2-bromoisobutyryl bromide (0.94 mL,
7.5 mmol) was added during 5 min. The colour changed from
deep black–brown to yellowish orange. On stirring for 1 h at
0 ЊC a precipitate (of salt) was formed. The reaction mixture
was poured into 25 mL 10% NH4Cl. The organic layer was
separated off and the aqueous layer extracted with 3 × 50 mL
Et2O. The combined organic layers were washed with 3 × 25 mL
water, 25 mL brine, dried (Na2SO4) and the solvent evaporated.
The main component (on TLC) of the residue was purified on a
silica column by elution with 0–15% Et2O in hexane to afford
0.75 g (2.7 mmol, 54%) light yellow solid, mp 112 ЊC (from
Et2O) (Found C, 77.76; H, 6.46; N, 10.02. C18H18N2O requires
C, 77.67; H, 6.52; N, 10.06 %); IR νmax: 3045–2819, 1708, 1618,
1576, 1565, 1476, 1458, 1446, 1322, 1241, 1094, 778, 702, 574
cmϪ1; δH: 1.32 (6 H, s), 2.74 (3 H, s), 7.15–7.17 (2 H, m), 7.30
(1 H, d, J 7.9), 7.48–7.55 (3H, m), 7.59–7.61 (3H, m); δC: 20.5
(q), 32.9 (q), 76.7 (s), 122.8 (d), 128.7 (d), 129.4 (d), 129.4 (d),
131.6 (d), 131.8 (s), 132.1 (d), 135.9 (s), 151.1 (s), 167.2 (s),
191.7 (s).
General procedure for preparation of 5.0 mmol 2-amino-
benzoimine dianions 2
To a slowly stirred solution of 10.8 mmol Grignard reagent in
10–20 mL THF or Et2O (THF for aryl reagents, Et2O for alkyl
reagents) the nitrile (5.0 mmol) was added in portions as a solid
(the condenser removed). CAUTION! May react violently! The
dianion was allowed to fully form during 15 min of reflux. To
form 2a, anthranilonitrile (0.59 g, 5.00 mmol) was treated with
10.8 mmol PhMgBr in 10–20 mL THF.
2-Aminobenzophenone imine (3a)
Compound 4b was also obtained by methylation of 4a: com-
pound 4a (1.32 g, 5.00 mmol) in 15 mL DMF was treated with
60% NaH (0.21 g, 5.3 mmol). To the dark orange solution
(formed after a few minutes) MeI (0.47 mL, 7.5 mmol) was
added, and the reaction mixture stirred for 30 min. The yellow
solution was poured into 60 mL water and left overnight,
followed by extraction with 3 × 50 mL Et2O. The combined
ether layers were washed with 3 × 25 mL water, 25 mL brine,
dried (Na2SO4) and the solvent evaporated to yield a yellow oil.
The oil was dissolved in 25 mL Et2O and filtered through a plug
of silica. Slow evaporation of the solvent left 0.49 g (1.8 mmol,
35%) light yellow crystalline material, which was identified as
4b (IR and TLC).
Dianion 2a (40 mmol from 4.72 g anthranilonitrile and PhMg-
Br) in 45 mL THF at 25 ЊC was quenched with 50 mL sat.
NH4Cl. The phases were separated and the aqueous phase
extracted with 2 × 50 mL Et2O. The combined organic phases
were washed with 2 × 20 mL water, 20 mL brine and dried
(Na2SO4). Evaporation of the solvent yielded an oil which
slowly solidified over several days at Ϫ15 ЊC. Hexane was added
and the solid cake crushed to small pieces. The solid material
was filtered off and washed with two portions of hexane to give
6.08 g (31 mmol, 77%) yellow crystalline material, mp. 45 ЊC
(lit.2 mp 48 ЊC from diisopropylether).
1,2-Dihydro-2,2-dimethyl-5-phenyl-3H-1,4-benzodiazepin-3-one
(4a) and 2-(1-Bromo-1-methylethyl)-4-phenylquinazoline (5a)
1,2-Dihydro-2,2-diethyl-5-phenyl-3H-1,4-benzodiazepin-3-one
To a 20 mL THF-solution of 5.0 mmol 2a, 2-bromoisobutyryl
bromide (0.94 mL, 7.5 mmol) was added in portions. The
reaction mixture was refluxed for 3 h and then poured into 25
mL 10% NH4Cl. The organic layer was separated off and the
aqueous layer extracted with 2 × 50 mL Et2O. The combined
organic layers were capped and left overnight, after which fine
needles could be filtered off. More material precipitated from
the filtrate on addition of hexane to a total of 0.93 g 4a (3.5
mmol, 70%) yellow fine needles, mp 197 ЊC. This compound
exists as two different conformers in the solid state, the
structure of which have previously been determined by X-ray
crystallography;1 IR νmax: 3299, 2991, 1682, 1612, 1596, 1571,
(4c)
To a 20 mL THF-solution of 5.0 mmol 2a, 2-bromo-2-ethylbu-
tyryl bromide (1.15 mL, 7.5 mmol) was added. The reaction
mixture was refluxed for 3 h and then poured into 25 mL 10%
NH4Cl. The organic layer was separated off and the aqueous
layer extracted with 2 × 50 mL Et2O. The combined organic
layers were washed with 50 mL water, 50 mL brine, dried
(Na2SO4) and evaporated to a yellow oil, which was purified
through a short silica column with hexane, hexane–diisopropyl
ether and finally Et2O. The final eluent portion contained the
product. Evaporation of the solvent afforded the product which
was recrystallised from diisopropyl ether to yield 0.16 g (0.55
mmol, 11%) yellow needles, mp 170 ЊC (Found C, 78.12; H,
6.85; N, 9.53. C19H20N2O requires C, 78.05; H 6.89; N 9.58%);
IR νmax: 3299, 2964, 2934, 1685, 1624, 1575, 1558, 1475, 1456,
1326, 1244, 756, 694, 637 cmϪ1; δH: 0.84 (6 H, t, J 7.3), 1.63 (4 H,
m), 6.71 (1 H, dt, J 0.7, 7.5), 6.93 (1 H, br s), 7.16 (2 H, m), 7.36
(1 H, dt, J 1.3, 7.7), 7.46–7.60 (5 H, m); δC: 7.6 (q), 25.0 (t), 69.8
1449, 1324, 1252, 1161, 1102, 954, 918, 769, 696, 640 cmϪ1
;
δH: 1.29 (6 H, s), 6.72 (1 H, t, J 7.4), 6.74–7.19 (2 H, m), 7.28
(1 H, br s), 7.38 (1 H, t, J 7.4), 7.50–7.61 (5 H, m); δC: 24.1 (q),
62.4 (s), 116.7 (d), 117.8 (s), 120.1 (d), 128.4 (d), 129.4 (d), 130.7
(d), 132.6 (d), 132.8 (d), 138.9 (s), 147.0 (s), 166.2 (s), 173.7 (s).
The remaining filtrate was filtered through a short silica plug
and concentrated. On addition of more hexane, a precipitate of
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 6 7 – 3 7 2
370

View Article Online
(s), 116.7 (d), 118.1 (s), 120.4 (d), 128.4 (d), 129.4 (d), 130.7 (d),
132.5 (d), 132.7 (d), 138.7 (s), 147.3 (s), 165.7 (s), 174.1 (s).
C, 81.33; H, 6.50; N, 11.78. C16H16N2 requires C, 81.32; H, 6.82;
N, 11.85%); IR νmax: 3258, 3055, 2957–2870, 1619, 1563, 1487,
1341, 1315, 1262, 1129, 753, 701 cmϪ1; δH: 1.03 (3 H, t, J 7.3),
1.79 (2 H, m), 4.70 (1 H, t, J 5.6), 6.34 (1 H, br s), 6.56 (1 H, t,
J 7.9), 6.74 (1 H, d, J 7.9), 6.94 (1 H, d, J 7.6), 7.19 (1 H, t,
J 7.6); δC: 9.0 (q), 28.7 (t), 70.0 (d), 114.0 (d), 116.2 (d), 116.3 (s),
127.6 (d), 128.0 (d), 128.7 (d), 129.0 (d), 132.3 (d), 138.2 (s),
148.3 (s), 163.9 (s).
2-(Chloromethyl)-4-phenylquinazoline (5b)
To a 10 mL THF solution of 5.0 mmol 2a, solid NH4Cl (0.27 g,
5.0 mmol) was added. The mixture was refluxed 5 min. The
flask was raised from the oil-bath and chloroacetyl chloride
(0.60 mL, 7.5 mmol) was added dropwise. After yet another
reflux period of 15 min, 50 mL water was added. After a few
minutes the liquid was decanted off and extracted with 2 × 20
mL EtOAc which was then added to the solid residue. The
solvent was evaporated. The red residue was recrystallised from
95% EtOH to yield 0.78 g (3.2 mmol, 64%) beige needles, mp
128 ЊC (Found C, 70.59; H, 4.30; N, 10.88. C15H11ClN2 requires
C, 70.73; H, 4.35; N, 11.00%); IR νmax: 3026, 2966, 1611, 1561,
1539, 1486, 1442, 1383, 1347, 1270, 840, 772, 700, 622 cmϪ1; δH:
5.00 (2 H, s), 7.52–7.65 (3 H, m), 7.78–7.81 (3 H, m), 8.08–8.11
(3 H, m); δC: 47.8 (t), 121.0 (s), 126.9 (d), 128.3 (d), 128.7 (d),
128.8 (d), 130.0 (d), 130.3 (d), 134.7 (d), 136.3 (s), 150.7 (s)
160.9 (s), 168.7 (s).
1,2-Dihydro-1-benzyl-4-ethyl-2-methylquinazoline (8b)
The nitrile 1a (1.04 g, 5.00 mmol) was added to 10.8 mmol
EtMgBr in 20 mL Et2O. The dianion was allowed to fully form
during 15 min of reflux. The dianion solution was cooled to rt
and acetaldehyde (0.56 mL, 10 mmol) was added dropwise. A
yellow precipitate formed. The mixture was refluxed 15 min,
and then quenched with 20 mL 10% NH4Cl. The layers were
separated and the aqueous layer extracted with 2 × 20 mL Et2O.
The combined organic layers were washed 2 × 20 mL water, 20
mL brine, dried (Na2SO4) and evaporated to give a yellow oil
(1.56 g). The part soluble in 60 mL boiling hexane was put on a
silica dry-flash column and eluted with 1 : 1 diisopropyl ether–
hexane followed by diisopropyl ether, and finally a large volume
of Et2O from which the product was recovered by evaporation
of the solvent. This yielded 0.59 g (2.2 mmol, 44%) yellow oil
(pure on TLC); HRMS found: [M ϩ H]ϩ, 265.1712 (C18H21N2
requires 265.1705); δH: 1.02–1.12 (6 H, m), 2.57–2.67 (2 H, m),
4.37 (1 H, d, J 16.3), 4.49 (1 H, d, J 16.3), 5.15 (1 H, q, J 6.0),
6.55 (1 H, d, J 8.1), 6.61 (1 H, dt, J 0.9 and 7.5), 7.15 (1 H, dt,
J 1.5 and 7.8), 7.23–7.38 (6 H, m); δC: 10.9 (q), 17.4 (q), 27.0 (t),
50.1 (t), 70.7 (d), 112.3 (d), 116.1 (d), 117.2 (s), 125.5 (d), 126.9
(d), 128.4 (d), 132.2 (d), 138.2 (s), 144.3 (s), 163.3 (s).
2-(Dichloromethyl)-4-phenylquinazoline (5c)
To a 20 mL THF solution of 5.0 mmol 2a, solid NH4Cl (0.54 g,
10 mmol) was added. The mixture was refluxed 5 min. The flask
was raised from the oil-bath and dichloroacetyl chloride (0.72
mL, 7.5 mmol) was added dropwise. After 30 min. the solvent
was evaporated in vacuo. The residue was dissolved in 10 mL
water and 40 mL 95% EtOH by heating. On slow cooling fine
needles formed and was filtered off to yield 0.46 g (1.6 mmol,
32%) off-white needles, mp 185 ЊC (lit.31 mp 185 ЊC); IR νmax
:
2997, 1611, 1567, 1537, 1487, 1385, 793, 771, 702, 685, 626
cmϪ1; δH: 7.55 (1 H, s), 7.66–7.68 (3 H, m), 7.84–7.86 (3 H, m),
8.15–8.17 (3 H, m); δC: 72.1 (d), 121.5 (s), 127.1 (d), 128.6 (d),
128.7 (d), 129.7 (d), 130.1 (d), 130.5 (d), 135.2 (d), 136.0 (s),
150.2 (s), 160.1 (s), 169.6 (s).
1,2-Dihydro-2-phenylquinazoline (9a)
To anthranilonitrile (0.59 g, 5.0 mmol) in 10 mL Et2O at 0 ЊC,
1 M DIBAL-H in toluene (10 mL, 10 mmol) was added under
nitrogen atm. After addition the mixture was stirred 1 h 0–25
ЊC. Benzaldehyde (1.6 mL, 15 mmol) was added dropwise and
stirring continued for 1 h. The reaction mixture, followed by a
rinsing portion of 20 mL Et2O, was poured into a separatory
funnel and washed with 2 × 20 mL 4 M NaOH. The combined
alkaline portions were in turn extracted with 20 mL Et2O. The
combined organic layers were washed with 20 mL water and
extracted with 3 × 20 mL 2 M HCl. The combined HCl
portions were washed with 20 mL hexane and then carefully
made basic with about 40 mL sat. Na2CO3. The colour changed
from red to pale yellow and the formed precipitate was filtered
off and dried. This afforded 0.48 g (2.3 mmol, 48%) of a pale
yellow solid, mp. 126 ЊC (lit.32 mp 131–132 ЊC from benzene–
diisopropyl ether); HRMS: found [M ϩ H]ϩ, 209.1082
(C14H12N2 requires 209.1079); IR νmax: 3227, 3053, 1630, 1569,
1479, 1455, 1269, 909, 743, 698 cmϪ1; δH: 5.99 (1 H, br s),
6.55–6.61 (2 H, m), 6.66 (1 H, s), 7.13–7.17 (2 H, m), 7.28–7.40
(3 H, m), 7.46–7.49 (2 H, m), 8.11 (1 H, br s); δC: 71.2 (d), 113.1
(d), 116.1 (s), 116.6 (d), 126.8 (d), 127.7 (d), 127.9 (d), 128.2 (d),
133.1 (d), 143.6 (s), 145.7 (s), 157.5 (d).
2-Bromo-N-methyl-N-(2-cyanophenyl)-2-methylpropanamide
(6b)
2-Bromoisobutyryl bromide (3.0 mL, 24 mmol) was added
dropwise to N-methylanthranilonitrile (1.32 g, 10.0 mmol) and
3.0 mL pyridine in 50 mL Et2O. After stirring of the reaction
mixture for 16 h, it was poured into 100 mL water. The layers
were separated and the aqueous layer extracted with 50 mL
Et2O. The combined organic layers were washed with 3 × 25 mL
water and 50 mL brine, dried (Na2SO4) and evaporated to yield
an oil which crystallised several hours after addition of hexane.
The solid was filtered off and washed with several portions of
hexane to yield 2.09 g (7.46 mmol, 75%) of a crystalline
material, mp 70 ЊC (from hexane–2-propanol) (Found C, 51.34;
H, 4.73; N, 9.95. C12H13BrN2O requires C, 51.26; H, 4.66; N,
9.96%); IR νmax: 3076–2925, 2230, 1644, 1596, 1486, 1449, 1359,
1092, 773, 749, 620, 553, 481 cmϪ1; δH: 1.88 (6 H, s), 3.45 (3 H,
s), 7.57 (1 H, t, J 7.8), 7.65 (1 H, d, J 7.9), 7.81 (1 H, dt, J 1.4
and 7.9), 7.94 (1 H, dd, J 1.2 and 7.8); δH: 32.1 (q), 40.7
(q, CH3), 57.8 (s), 116.2 (s), 128.7 (d), 129.4 (d), 133.6 (d), 134.5
(d), 146.7 (s), 169.5 (s)
1,2-Dihydro-2-ethyl-4-phenylquinazoline (8a)
1,2-Dihydro-6-bromo-2-phenylquinazoline (9b)
To 4.0 mmol of dianion 2a in 10 mL THF, propionaldehyde
(0.57 mL, 8.0 mmol) was added, and the reaction mixture was
refluxed 2 h. After cooling to rt, 10 mL 10% NH4Cl was added,
followed by extraction with 3 × 20 mL Et2O. The combined
ethereal layers were washed with 3 × 10 mL water and 20 mL
brine, dried (Na2SO4) and evaporated to a yellow oil. The
yellow product was purified by column chromatography on
silica with 0–25% Et2O in hexane to yield 0.45 g semisolid
material which could be crystallised from hexane to yield 0.30 g
(1.3 mmol, 33%) yellow crystalline material, mp 99.5 ЊC (Found
2-Amino-5-bromobenzonitrile (0.99 g, 5.00 mmol; prepared by
bromination of 2-aminobenzonitrile in AcOH)33 treated with
DIBAL-H according to the same procedure as reported for
10a, yielded 0.54 g (1.9 mmol, 38%) yellow solid, mp. 129 ЊC;
HRMS: found [M ϩ H]ϩ, 287.0188 (C14H11BrN2 requires
287.0184). IR νmax: 3235, 1628, 1562, 1478, 1454, 1266, 1183,
820, 757, 700 cmϪ1; δH: 6.05 (1 H, br s), 6.55 (1 H, d, J 8.3), 6.88
(1 H, s), 7.26–7.47 (7 H, m), 8.12 (1 H, br s); δC: 71.1 (d), 106.3
(s), 106.8 (s), 115.2 (d), 126.7 (d), 127.0 (d), 128.1 (d), 130.0 (d),
135.4 (d), 143.2 (s), 144.7 (s), 156.3 (d).
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 6 7 – 3 7 2
371

View Article Online
7 W. Szczepankiewicz, J. Suwinski and R. Bujok, Tetrahedron, 2000,
56, 9343–9349.
Spiro[cyclohexane-1,2Ј-(1H)-4Ј-phenylquinazoline] (10a)
To 5.0 mmol dianion 2a in 20 mL THF, solid NH4Cl (0.54 g, 10
mmol) was added, and the mixture was refluxed 5 min. Cyclo-
hexanone (1.55 mL, 15 mmol) was added and the reflux con-
tinued for 19 h. 10 mL 10% NH4Cl and 20 mL Et2O was added
and the phases separated. The aqueous phase was extracted
with 2 × 20 mL Et2O and the combined organic phases were
washed with 4 × 20 mL water. The washed solution was in turn
extracted with 4 × 20 mL 5% citric acid. After the combined
acidic portions were neutralised with about 50 mL sat. Na2CO3,
a solid slowly (20 h) formed and was filtered of. After washing
with water and drying, 0.97 g (3.5 mmol, 70%) of a mustard
coloured solid remained, mp 143 ЊC (lit.16 146 ЊC from
diisopropyl ether); IR νmax: 3288, 2927, 2846, 1610, 1559, 1474,
1444, 1325, 746, 700 cmϪ1; δH: 1.47–1.87 (10 H, m), 6.30 (1 H, br
s), 6.52 (1 H, t, J 7.3); δC: 21.1 (t), 25.4 (t), 37.2 (t), 69.3 (s), 114.3
(d), 115.5 (s), 115.7 (d), 127.4 (d), 128.0 (d), 128.7 (d), 128.9 (d),
132.3 (d), 138.5 (s), 146.4 (s), 161.7 (s).
8 J. A. Seijas, M. P. Vazquez-Tato and M. M. Martinez, Tetrahedron
Lett., 2000, 41, 2215–2217.
9 F. Sauter, J. Froehlich, K. Blasl and K. Gewald, Heterocycles, 1995,
40, 851–866.
10 J. B. Hynes, A. Tomazic, A. Kumar, V. Kumar and J. H. Freisheim,
J. Heterocycl. Chem., 1991, 28, 1981–1986.
11 T. Mizuno, N. Okamoto, T. Ito and T. Miyata, Heteroat. Chem.,
2000, 11, 428–433.
12 T. Mizuno, N. Okamoto, T. Ito and T. Miyata, Kagaku to Kogyo
(Osaka), 2000, 74, 149–151.
13 T. Miyata, T. Mizuno, Y. Nagahama, I. Nishiguchi, T. Hirashima
and N. Sonoda, Heteroat. Chem., 1991, 4, 473–475.
14 M. T. Chhabria and C. J. Shishoo, Heterocycles, 1999, 51,
2723–2729.
15 J. Bergman, A. Brynolf and E. Vuorinen, Tetrahedron, 1986, 42,
3689–3696.
16 J. Bergman, A. Brynolf, B. Elman and E. Vuorinen, Tetrahedron,
1986, 42, 3697–3706.
17 L. Strekowski, M. T. Cegla, D. B. Harden, J. L. Mokrosz and
M. J. Mokrosz, Tetrahedron Lett., 1988, 29, 4265–4268.
18 A. Bischler and M. Lang, Chem. Ber., 1895, 28, 279–293.
19 T. L. Gilchrist, C. J. Harris, C. J. Moody and C. W. Rees, J. Chem.
Soc., Perkin Trans. 1, 1975, 19, 1969–1972.
1,2-Dihydro-2,2-dimethyl-4-phenylquinazoline (10b)
Following the procedure for compound 11a, acetone (1.5 mL, 20
mmol) was added to 5.0 mmol 2a and the reflux continued for 3
h. The same work-up procedure yielded 0.78 g (3.3 mmol, 66%)
large yellow crystals, mp 115 ЊC (from heptane) (Found C, 81.31;
H, 6.76; N, 11.91. C16H16N2 requires C, 81.32; H, 6.82; N,
11.85%); IR νmax: 3299, 2976, 1623, 1569, 1538, 1329, 1286, 1195,
1168, 958, 745, 696, 677 cmϪ1; δH: 1.39 (6 H, s), 6.42 (1 H, br s),
6.50 (1 H, dt, J 0.8 and 7.5), 6.65 (1 H, dd, J 0.8 and 8.3), 6.90 (1
H, d, J 7.5), 7.16 (1 H, dt, J 1.5 and 7.5), 7.45 (5 H, m); δC: 28.6
(q), 68.6 (s), 114.0 (d), 114.8 (s), 115.7 (d), 127.4 (d), 128.0 (d),
128.6 (d), 128.9 (d), 132.4 (d), 138.2 (s), 146.7 (s), 161.6 (s).
20 L. Strekowski, M. T. Cegla, S. B. Kong and D. B. Harden,
J. Heterocycl. Chem., 1989, 26, 923–928.
21 Q. Ann, M. Ou, R. Madigan and J. Adams, Org. Mass Spectrom.,
1992, 27, 916–918.
22 J. P. Michael, Natural Product Reports, 2001, 18, 543–559.
23 D. T. Hurst, Rodd’s Chemistry of Carbon Compounds (2nd Edition),
2000, 4, 71–98.
24 S. Johne, Rodd’s Chemistry of Carbon Compounds (2nd Edition),
2000, 4, 203–231.
25 S. Johne, Alkaloids (Academic Press), 1986, 29, 99–140.
26 C. Melchiorre, P. Angeli, M. L. Bolognesi, A. Chiarini, D. Giardina,
U. Gulini, A. Leonardi, G. Marucci, A. Minarini, M. Pigini,
W. Quaglia, M. Rosini and V. Tumiatti, Pharm. Acta Helv., 2000,
74, 181–190.
27 J. L. Reid and J. Vincent, Cardiology, 1986, 73, 164–174.
28 D. M. Patel, A. J. Visalli, J. J. Zalipsky and N. H. Reavey-Cantwell,
Anal. Profiles Drug Subst., 1975, 4, 245–267.
29 H. Quast, S. Ivanova, E.-M. Peters and K. Peters, Eur. J. Org. Chem.,
2000, 8, 1577–1587.
30 E. M. Campi, W. R. Jackson, Q. J. McCubbin and A. E. Trnacek,
Aust. J. Chem., 1996, 49, 219–230.
References
1 J. Bergman, A. Brynolf, K. W. Toernroos, B. Karlsson and
P. E. Werner, Heterocycles, 1983, 20, 2145–2148.
2 J. Schmitt, M. Suquet, J. Boitard, P. Comoy, J. P. Meingan, G. Gallet,
T. Clim, C. Perrin and J. Le Meur, Ind. Chim. Belge, 1967, 32,
184–187.
3 J. B. Culbertson, J. Am. Chem. Soc., 1951, 73, 4818–4823.
4 J. B. Culbertson, D. Butterfield, O. Kolewe and R. Shaw, J. Org.
Chem., 1962, 27, 729–733.
5 F. J. Weiberth and S. S. Hall, J. Org. Chem., 1987, 52, 3901–3904.
6 A. Couture, E. Deniau, Y. Gimbert and P. Grandclaudon,
Tetrahedron, 1993, 49, 1431–1444.
31 H. Meerwein, P. Laasch, R. Mersch and J. Nentwig, Chem. Ber.,
1956, 89, 224–238.
32 B. A. Tertov, P. P. Onishchenko and V. V. Bessonov, Chem.
Heterocycl. Compd., 1974, 10, 1237–1240.
33 S. M. Mackenzie and M. F. Stevens, J. Chem. Soc. C, 1970, 17,
2298–2308.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 3 6 7 – 3 7 2
372