Synthesis of unsymmetrical 3,3′-biquinazoline-2,2′-diones by condensation of 3-aminoquinazolinones with benzoxazinones; fortuitous discovery, and further syntheses of 4-H-3-oxo-1,9a,10-triazaanthracen-9-ones

Article abstract of DOI:10.1039/b419108k

Condensation of 2-alkyl- or 2-aryl-3-aminoquinazolin-4-ones with benz[1,3]oxazin-4-ones gives the unsymmetrical 2,2′ disubstituted 3,3′ biquinazoline-4,4′-diones. The reaction is tolerant to a range of heteroatom and unsaturated functionality in the quina

Full text of DOI:10.1039/b419108k

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Synthesis of unsymmetrical 3,3ꢀ-biquinazoline-2,2ꢀ-diones by
condensation of 3-aminoquinazolinones with benzoxazinones;
fortuitous discovery, and further syntheses of 4-H-3-oxo-1,9a,10-
triazaanthracen-9-ones
Michael P. Coogan,* Li-ling Ooi and Fabrizio Pertusati
Department of Chemistry, Cardiff University, Park Place, Cardiff, Wales (Cymru), CF10 3TB,
UK. E-mail: cooganmp@cf.ac.uk; Fax: 02920 874030; Tel: 02920 874066
Received 21st December 2004, Accepted 4th February 2005
First published as an Advance Article on the web 22nd February 2005
Condensation of 2-alkyl- or 2-aryl-3-aminoquinazolin-4-ones with benz[1,3]oxazin-4-ones gives the unsymmetrical
2,2ꢀdisubstituted 3,3ꢀbiquinazoline-4,4ꢀ-diones. The reaction is tolerant to a range of heteroatom and unsaturated
functionality in the quinazolinone 2-position. However, treatment of 3-amino-2-hydroxymethyl-3H-quinazolin-4-
ones with benz[1,3]oxazinone at high temperatures gave 4H-3-oxo-1,9a,10-triazaanthracen-9-ones, an unreported
fused heterocyclic system, a more direct synthesis of which, replacing benzoxazinones with orthoesters is presented.
3,3ꢀ-biquinazoline-4,4ꢀ-diones, and seemed a feasible† route by
Introduction
analogy with the reliable route to 3-aminoquinazolinones via the
treatment of benz[1,3]oxazin-4-ones with hydrazines.3 As a large
range of 2-substitution patterns are known and readily available
both for 3-aminoquinazolin-4-ones and for benz[1,3]oxazin-4-
ones, thus, at least in theory, a wide range of 3,3ꢀ-biquinazoline-
4,4ꢀ-diones should be available from this specific heterocoupling
reaction (Scheme 3).
This group is interested in the preparation, and application as
ligands and in asymmetric synthesis, of 3,3ꢀ-biquinazoline-4,4ꢀ-
diones,1 a family of axially chiral bis-heterocycles which have
high barriers to rotation around an N–N bond and thus form sta-
ble atropisomers. Our previous reports have demonstrated that
the syntheses of C2 symmetrical 3,3ꢀ-biquinazolinone are readily
achieved from bisanthranoyl hydrazine, (Scheme 1) however this
route seems unpromising for the synthesis of unsymmetrical
2,2ꢀ disubstituted 3,3ꢀ-biquinazolinones, and thus other routes
were explored which would allow the synthesis of each quina-
zolinone unit separately followed by the coupling of the two
halves.
Scheme 3
Results and discussion
Synthesis of 3,3ꢀ-biquinazoline-4,4ꢀ-diones via reaction of
3-aminoquinazolin-4-ones and benz[1,3]oxazin-4-ones
Scheme 1
To investigate the viability of this synthesis a range of 3-
aminoquinazolinones (1–5)4–7 were prepared, both known com-
pounds by literature methods or adaptations, with a range
of substituents of various steric bulks, and reactivity pro-
files, including heteroatoms, unsaturation and free hydroxyl
groups. Three examples of the other partner in the cou-
pling were prepared, the parent 2-protio-8, 2-methyl9 and 2-
ethylbenz[1,3]oxazin-4-one10 in order to test the effect on the
reaction of increased steric demands in this partner. The initial
attempts at condensation were run under conditions which
mimicked those used in the synthesis of 3-aminoquinazolin-4-
ones from the reaction of benz[1,3]oxazinones with hydrazine,3
that is reflux in alcoholic solvents, but it soon became clear that,
especially in the case of the unsubstituted benz[1,3]oxazin-4-one,
alcoholysis of the oxazine ring dominates the reaction. Presum-
ably, this is a reflection of the much diminished nucleopilicity
of 3-aminoquinazolin-4-ones when compared with hydrazine
itself, again reflecting the electron-withdrawing nature of the
Although an intramolecular oxidative coupling of quinazoli-
none fragments has been reported2 (Scheme 2) this reaction,
if applied to the intermolecular coupling of unsymmetrically
substituted quinazolinones, would be expected to give statistical
mixtures of products of homo- and hetero-coupling. Therefore a
heterocoupling route seemed indicated, in which the two halves
of a non-symmetrical biquinazolinone could be independently
prepared and reliably coupled in a controlled manner to give the
unsymmetrical product.
Scheme 2
One promising avenue seemed to be the condensation of 3-
aminoquinazolinones with benz[1,3]oxazin-4-ones, which would
allow the synthesis of both symmetrical and unsymmetrical
† We thank a referee to drawing our attention to other precedent16 for
this approach of which we were unaware.
1 1 3 4
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 3 4 – 1 1 3 9
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 5
©

View Article Online
Table 1
quinazolin-4-one ring. Refluxing toluene or other inert solvents
gave condensation, but with poor yields and unacceptably slow
rates (reaction times of many days or weeks) however the
addition of catalytic p-toluene sulfonic acid accelerated the
reaction to such an extent that, typically, moderate to good
yields could be obtained in a few hours to a day reaction
time under Dean–Stark conditions. In most cases the reaction
proceeded smoothly to give the 3,3ꢀ-biquinazoline-4,4ꢀ-dione as
the only major product and as a crystalline solid, isolable at
analytical purity by simple crystallisation of the crude reaction
mixtures without the need for extractions, separations or chro-
matography. The limitations of this synthesis of unsymmetrical
2,2ꢀ-substituted 3,3ꢀ-biquinazoline-4,4ꢀ-diones initially appeared
to be essentially steric with heteroatom and unsaturated sub-
stitution tolerated. While 2-tert-butyl-3-aminoquinazolin-4-one
condenses efficiently with 2-unsubstituted benzoxazinone, it fails
to give any 3,3ꢀbiquinazoline-4,4ꢀ-dione with the methyl or ethyl
analogues, only the open amide being observed in the reaction
mixture, no matter how forcing the conditions used for the
reaction, even when strong dehydrating agents such as acetic
anhydride or even thionyl chloride were employed at reflux,
eventually giving decomposition of the intermediate without
providing the ring-closed product. These open amide products
proved to be unstable and, in cases where the cyclisation to
the desired product failed, were never isolated as homogeneous
substances. It is non-intuitive that the sterically hindered 2-tert-
butyl-3-aminoquinazolinone should react readily with all the
benzoxazinones, and only demonstrate a steric problem at the
cyclisation stage, but this can be rationalised in terms of the
planarity of the benzoxazinone electrophile. Initial attack at the
benzoxazinone carbonyl will suffer from only slight steric clashes
with the quinazolinone 2-substituent, and the benzoxazinone
side chain will be almost irrelevant, however in the cyclisation
step the two substituents are forced into proximity, and thus it
is at this stage that the bulk of the tert-butyl group becomes
too demanding a match for all but the smallest benzoxazinone
(Scheme 4).
Entry QNH2, R2
R1
% cat/h BiQ Yield (%)
1
2
3
4
5
6
1 Ph4
6 Et10 5/10
6 Et 5/10
9
10
11
12
13
14
79
44
62
40
51
47
2 SEt (this paper)
1 Ph4
7 Me9 5/6
3 tBu5
8 H8
6 Et
8/8
8/24
5/24
4 CO2Et6
7
=
5 CHMeCH2CH CHPh 8 H
the presence of either heteroatoms or unsaturation in the 2-
substituent of the quinazolin-4-one which would be important
in attempts to apply such products as ligands for metals in
asymmetric catalysis, or as chiral auxiliaries in which faces of a
double bond could be rendered diastereotopic by the chiral axis.
Attempted synthesis of chiral, non-racemic biquinazolinones
All of the examples of biquinazolinones synthesised by this
coupling reaction above are racemic, only 5 being chiral and
itself a racemate, and although we have previously demon-
strated deracemisation of a small number of examples of 3,3ꢀ-
biquinazolinones1 attempts to apply this protocol to other
examples have proved less successful. Thus a synthesis of a
biquinazolinone in non-racemic form, or at least bearing a non-
racemic chiral centre which would assist with the resolution of
the chiral axis was desirable. Incorporation of a chiral centre into
the 3-position of a 3-aminoquinazolinone seemed the simplest
approach, as such compounds are well known in the literature.
It was hoped that a single chiral centre in the 2-alkyl substituent
may be sufficient to give good control of the formation of
the chiral axis, and it seemed likely that the best control may
come from a chiral group containing a hydroxyl substituent
which could hydrogen bond to the amide intermediate in the
condensation. Thus a sample of 2-(1-hydroxy-2-methyl)propyl)-
3-aminoquinazolin-4-one 15 (prepared from L-valine by the
procedure of Atkinson11) was treated with benz[1,3]oxazinone
8 in toluene at reflux with a catalytic quantity (5 mol%) of
toluene-4-sulfonic acid under Dean–Stark conditions. None of
the expected biquinazolinone was recovered, instead a highly
crystalline compound of molecular weight m/z = 244 was
recovered in 49% yield after cooling to room temperature and
crystallisation of the precipitate from methanol (Scheme 6).
Scheme 6
Proton and carbon NMR suggested that the structure was
derived from the 3-aminoquinazolinone with the addition of
an additional methine which, from the chemical shift data
(dH 7.2, dC 126) appeared to be heterocyclic, which, along with
the loss of all exchangeable protons, suggested the structure 16.
This structure was eventually confirmed by single crystal X-ray
diffraction measurements on crystals grown by slow evaporation
of a methanol solution (Fig. 1).
Scheme 4
Other than this steric limitation a number of examples
of condensation of each benz[1,3]oxazine with various 3-
aminoquinazolin-4-ones were demonstrated (see Scheme 5,
Table 1) and the reaction did not seem to be unduly effected by
Compound 16 was observed to have crystallised in space
group P21/c which is not a chiral space group, with the
asymmetric unit being repeated through an improper rotation,
which is impossible for crystals containing homochiral material.
This indicated that the chiral centre of the valine-derived
substituent of the quinazolinone had racemised at some point
along the reaction scheme (confirmed for the bulk sample by
the zero optical rotation). This racemisation was unexpected as
such chiral carbons are usually stable to the reaction conditions
involved in this cyclisation, and subsequently it emerged that
Scheme 5
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 3 4 – 1 1 3 9
1 1 3 5

View Article Online
Table 2
Entry
Q, R1
R2
Time/h
Product
Yield (%)
1
2
3
4
5
6
15, iPr
15, iPr
15, iPr
17, Ph
19, H
H
5
24
36
24
24
36
16
21
22
18
20
23
68
77
54
58
85
60
Me
Et
H
H
Ph
15, iPr
aminoquinazolinone, and not a function of the new heterocycle
synthesis. The mandelic acid-derived 3-aminoquinazolinone 17
also reacted with triethyl orthoformate to give the analogous
heterocycle 18 in enantiomerically pure form and with a 60%
yield. The simplest 2-hydroxymethyl-3-aminoquinazoline 1912
also condensed smoothly with triethyl orthoformate to give
the parent oxatriazaanthacenone 20 in 85% yield. All of these
compounds were isolated by crystallisation of the reaction
mixtures after evaporation of excess orthoformate without the
need for chromatography.
Although orthoformates are common starting materials it
was felt that other reagents, e.g anhydrides or acyl chlorides
may be better sources of the products which might be expected
from reaction with orthoacetates and higher analogues as these
are also electrophilic acyl equivalents, and are more common
and more reactive. Unfortunately, while the treatment of 3-
amino-2-[(S)-1-hydroxy-2-methylpropyl]quinazolin-4-one with
either acetic anhydride or acetyl chloride gave some of the
desired product, derived from N-acylation followed by carbonyl
activation, nucleophilic attack of the hydroxyl and loss of car-
boxylic acid, the major products in each case were the uncyclised
N-acetyl hydroxyl compound13 or the N,O-diacetyl compound,
depending upon conditions (Scheme 8). Similar results were
obtained using propionyl chloride and benzoyl chloride.
Fig. 1 ORTEP view of 16 50% probabilities.
the starting material 15 was itself racemic. As 15 is derived
from L-valine it was disturbing that racemisation had occurred,
as the diazotization–substituion reaction is a classic double
inversion and should proceed with retention of configuration. A
reinvestigation of the synthetic pathway to 15 demonstrated that
the diazotization step is particularly sensitive to heat and in the
absence of efficient cooling to between 0 and 5 ◦C racemisation
will occur as the reaction is also exothermic. It was also noted
that cooling to zero lowers the yield due to solidification of the
sodium nitrite–acetic acid mixture so a balance is necessary in
order to avoid racemisation but maintain high yields.
A simple mechanistic rational suggests that the benz[1,3]-
oxazin-4-one is acting as a formate synthon, with loss of
anthranilic acid, suggesting that the new heterocyclic system
could be accessed via the treatment of 2-hydroxymethyl-3-
aminoquinazolinones with orthoesters, and indeed it transpired
that treatment of 3-aminoquinazolinone 15 with triethyl or-
thoformate at reflux also gave oxotriazaanthracenone 16 in
a better yield of 68%. Initial attack of the quinazolinone
nitrogen at the activated carbonyl equivalent obtained from
triethyl orthoformate upon protonation–loss of ethanol to
give an orthoformamide (no strong acid is required, in fact
the reaction occurs without addition of external acid simply
upon heating the reactants) followed by another protonation–
loss of ethanol sequence gives the activated intermediate for
nucleophilic attack by the hydroxymethyl oxygen. The cyclic
orthoamide thus formed spontaneously loses ethanol to give the
highly conjugated oxatriazaanthracenone.
Synthesis of 4-H-3-oxo-1,9a,10-triazaanthracen-9-ones
Surprisingly, not only was the product unreported, but we
could find no reports of this heterocyclic system, or even the
bicyclic non-benzanullated analogue, in the literature. As routes
to new heterocyclic systems are continuously of interest, an
investigation of the synthesis of a small number of analogues
was undertaken. Repeating the reaction with a samples of enan-
tiomerically pure quinzolinones (see Scheme 7, Table 2) gave
the corresponding enantiomerically pure products with optical
rotation unchanged upon recrystallisation demonstrating that
the racemisation was solely associated with the synthesis of the 3-
Scheme 8
Turning attention back to the reactions of 3-amino-2-
hydroxymethylquinazolin-4-ones with orthoesters allowed ac-
cess to higher analogues of the new heterocycles (Scheme 8,
Table 2). Again all of these compounds were isolated in
homogeneous form from the crystallisation of the crude residue
obtained upon evaporation of the reaction mixture without the
need for chromatography.
Conclusions
A route to 2,2ꢀ-unsymmetrically substituted 3,3ꢀ-biquinazoline-
4,4ꢀ-diones has been developed coupling 3-aminoquinazolin-
4-ones and benz[1,3]oxazinones. After an unexpected product
was isolated from one of these attempted coupling reactions a
synthesis of a new heterocyclic system the 4-H-3-oxo-1,9a,10-
triazaanthracen-9-ones was developed.
Scheme 7
1 1 3 6
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 3 4 – 1 1 3 9

View Article Online
2-Methyl-2ꢀ-phenyl-3,3ꢀ-bisquinazoline-4,4ꢀ-one 11. A mix-
ture of 2-methyl-4H-3,1-benzoxazin-4-one (0.628 g, 4.2 mmol),
3-amino-2-phenyl-3H-quinazolin-4-one (1 g, 4.26 mmol), and
p-toluenesulfonic acid (0.047 g, 0.211 mmol) was heated at reflux
in toluene for 6 hours. After cooling, the toluene was evaporated
in vacuo and the solid obtained was crystallised from ethanol to
Experimental
General experimental
All starting materials and reagents were purchased from com-
mercial suppliers and used as supplied unless otherwise stated.
Benzoxazinones were synthesised by literature routes8–10 or
similar procedures; 3-aminoquinazolinones were synthesised by
literature routes4–7,11 unless described below. NMR spectra were
recorded on a Bruker DPX 400 MHz at 400 MHz (proton) and
160 MHz (carbon), or an APX 250 at 250 MHz and 100 MHz
respectively. IR spectra were recorded on a Perkin-Elmer 1600
FT IR as thin films or nujol mulls; mass spectra were recorded
on a VG Fisons Platform II or at the EPSRC National Mass
Spectrometry Service in Swansea (HRMS). Elemental analyses
were performed by Warwick Analytical Services (University of
Warwick).
◦
give white crystals mp 162 C (0.955 g, 61.5%). dH (400 MHz,
CDCl3) 8.29 (1H, dd, J = 8.0, 0.9 Hz, H5), 8.14 (1H, dd, J =
8.0, 1.4 Hz, H5ꢀ), 7.82 (2H, m), 7.69 (1H, ddd, J = 8.0 7.8
1.2 Hz H7ꢀ), 7.52 (M, 5H), 7.39 (1H, ddd, J = 8.0, 7.0, 1.2, H7),
=
7.28 (2H, m), 2.32 (1H, s, N C–CH3); dC (100 MHz, CDCl3)
159.8, 159.4, 154.9, 153.6, 147.2, 146.9, 136.2, 135.8, 132.7,
131.3, 129.0, 128.8, 128.4, 128.1, 128.0, 127.8, 127.7, 127.6,
121.4, 120.1, 22.0; mmax/cm−1 1692, 1599, 1566, 1378, 1324, 1268,
776, 698; m/z (APCI) (%) 381 (100) [M + H+]; C23H16N4O2
requires, C 72.62; H 4.24; N 14.73; found C 72.50; H 4.24;
N 14.50%.
2-tert-Butyl-3,3ꢀ-bisquinazoline-4,4ꢀ-dione 12. A mixture of
2-Ethylthio-3-aminoquinazolin-4-one 2. To a stirred solution
of 2-mercapto-3-aminoquinazolin-4-one14 (8.3 g, 0.043 mol) in
saturated aqueous sodium hydroxide (100 ml) was added ethanol
(25 ml) followed by ethyl iodide (20 ml, 39 g, 0.25 mol). The
solution was stirred for a further 1 h then a white solid which
had precipitated was collected by filtration. Crystallisation from
ethanol gave a white solid (6.6 g, 69%); mp 138–140 ◦C (ethanol).
dH 8.12 (1H, d J, = 6.2 Hz, H-5), 7.67 (3H, m), 4.85 (2H, br s,
NH2), 3.12 (2H, q, J = 9.0 Hz, SCH2), 1.33 (3H, t, J = 9.0 Hz,
SCH2CH3); dC 161.4, 158.6, 147.6, 134.3, 126.6, 126.4, 125.4,
118.3, 25.6, 13.9; mmax/cm−1 3320, 3220, 1680, 1630; m/z(%)
(EI) 221 (45)(M+) 205(68), 193(47), 174(11), 162(100), 144(69);
C10H11N3OS requires C 54.3; H 5.0; N 19.0%, found C 54.25; H
5.1; N 18.9%.
4H-3,1-benzoxazin-4-one (1.016 g,
7 mmol), 3-amino-2-
tertbutyl-3H-quinazolin-4-one (1 g, 4.6 mmol) and p-
toluenesulfonic acid (0.066 g, 0.345 mmol) was heated at reflux
with a Dean–Stark apparatus in toluene for 8 hours. After
cooling, the toluene was evaporated in vacuo and the solid
obtained was crystallised from ethanol to give pale brown
◦
crystals; mp 225 C (0.64 g, 38%); dH (400 MHz, CDCl3) 8.28
(1H, dd J = 7.9, 1.3 Hz, H4), 8.19 (1H, dd J = 7.7, 1.1 Hz,
H4ꢀ), 7.80 (1H, s, H2ꢀ), 7.75 (4H, m), 7.52 (1H, ddd J = 8.0, 6.7,
1.1 Hz, H6ꢀ), 7.44 (1H, ddd, J = 7.2, 6.1, 1.3 Hz, H6), 1.3 (9H, s,
C(CH3)3); dC (100 MHz, CDCl3), 159.63, 159.06, 158.83, 146.74,
146.09, 145.21, 134.49, 134.40, 127.33, 127.10, 127.09, 126.64,
126.48, 126.20, 121.52, 119.42, 38.71, 28.84; mmax/cm−1 1692,
1594; m/z (APCI)(%) 347 (100) [M + H+]; C20H18N4O2 requires
C 69.35, H 5.24, N 16.17; found C 68.95, H 5.24, N 15.95%.
2-Ethyl-2ꢀ-phenyl-3,3ꢀ-bisquinazoline-4,4ꢀ-one 9. A mixture
of 2-ethyl-4H-3,1-benzoxazin-4-one (3 g, 17.1 mmol), 3-
amino-2-phenyl-3H-quinazolin-4-one (3 g, 12.6 mmol) and p-
toluenesulfonic acid (0.147 g, 0.855 mmol) was heated at reflux in
toluene for 10 hours. After cooling, the toluene was evaporated
in vacuo and the solid obtained was triturated with diethyl ether
to give a pale yellow solid which was then crystallised from
ethanol to give white crystals; mp 94 ◦C (3.90 g, 78.6%). dH
(400 MHz, CDCl3) 8.30 (1H, dd J = 8.0, 1.3 Hz, H5), 8.15 (1H,
dd J = 7.9, 1.3 Hz, H5ꢀ), 7.83 (2H, m), 7.68 (1H, ddd, J = 8.0, 7.0,
1.3 Hz), 7.54 (m, 5H), 7.38 (ddd, 1H, J = 8.0, 7.1, 0.9 Hz), 7.25
2-Ethyl-2ꢀ-ethoxycarbonyl-3,3ꢀ-bisquinazoline-4,4ꢀ-dione
mixture of 2-ethyl-4H-3,1-benzoxazin-4-one (1.35 g,
13.
A
7.85 mmol), 3-amino-2-ethoxycarbonyl-3H-quinazolin-4-one
(1.220 g, 5.2 mmol) and p-toluenesulfonic acid (0.0746 g,
0.392 mmol) was heated at reflux in toluene for 24 hours. After
cooling, the toluene was evaporated in vacuo and the solid
obtained was triturated with diethyl ether to give a pale yellow
solid which was then crystallised from dichloromethane–light
petrol to give white crystals mp 175 ◦C (1.04 g, 50.9%). dH
(400 MHz, CDCl3) 8.30 (1H, dd, J = 7.9, 1.0 Hz, H5), 8.15
(1H, dd J = 7.6, 1.1 Hz, H5ꢀ), 7.86 (2H, m), 7.74 (2H, m), 7.60
(1H, ddd, J = 7.3, 7.2, 1.0 Hz, H7), 7.42 (1H, ddd, J = 7.3, 7.1,
=
(2H, m), 2.50 (2H, q, J = 7.3 Hz, N C–CH2CH3), 1.21 (3H,
t, J = 7.3 Hz, CH2CH3); dH (100 MHz, CDCl3) 160.0, 159.5,
156.5, 155.1, 147.0, 136.1, 135,7, 135.3, 132.7, 131.9, 131.2,
129.0, 128.7, 128.4, 128.0, 128.0, 127.8, 127.5, 121.5, 120.9, 26.6,
10.3; mmax/cm−1 1686, 1591, 1464, 1377, 1331, 1274, 1116, 909,
873, 764, 695; m/z(APCI)(%) 395.1 (100) [M + H+]; C24H18N4O2
requires C 73.08; H 4.60; N 14.20%, found C 73.01; H 4.62; N
14.12 (%).
1
1.1 Hz, H7ꢀ), 4.24 (1H, dq J = 11.2, 7.0 Hz, × CO2CH ),
2
2
1
2
4.19 (1H, dq J = 11.2, 7.0 Hz, × CO2CH ), 2.77 (1H, dq, J =
2
1
2
=
17.2, 7.2 Hz, × N C–CH ), 2.58 (1H, dq, J = 17.2, 7.2 Hz,
2
1
2
=
× N C–CH ), 1.31 (3H, t, J = 7.0 Hz, CO2CH2CH3), 1.13
2
=
(3H, t, J = 7.2 Hz, N C–CH2CH3); dC (100 MHz, CDCl3),
174.1, 158.16, 157.62, 155.90, 145.82, 144.45, 143.43, 134.87,
134.32, 128.47, 128.15, 126.86, 126.73, 126.20, 125.96, 121.42,
119.56, 62.64, 24.95, 12.65, 8.74; mmax/cm−1 1738, 1709, 1686,
1601; m/z (APCI) (%) 391 [M + H+]; C21H18N4O4 requires C
64.61; H 4.65; N 14.35, found C 64.54; H 4.64; N 14.29%.
2-Ethyl-2ꢀ-thioethyl-3,3ꢀ-bisquinazoline-4,4ꢀ-dione 10. A mix-
ture of 2-ethyl-4H-3,1-benzoxazin-4-one (0.63 g, 3.57 mmol),
3-amino-2-thioethyl-3H-quinazolin-4-one (0.71 g, 3.58 mmol)
and p-toluenesulfonic acid (0.030 g, 0.179 mmol) was heated
at reflux in toluene for 10 hours. After cooling, the toluene
was evaporated in vacuo and the solid obtained was triturated
with diethyl ether to give a pale yellow solid which was then
crystallised from dichloromethane–light petrol to give white
crystals mp 173 ◦C (0.600 g, 44.0%). dH (400 MHz, CDCl3)
8.22 (1H, dd, J = 7.8, 1.3 Hz, H5ꢀ), 8.17 (1H, dd, J = 7.9,
1.3 Hz, H5), 7.75 (3H, m), 7.61 (1H, d, J = 7.9 Hz, H8ꢀ), 7.44
(2H, m), 3.15 (2H, m), 2.54 (2H, m), 1.28 (6H, m); dC (100 MHz,
CDCl3) 158.8, 158.5, 156.6, 156.3, 147.4, 146.82, 135.6, 135.4,
127.8, 127.7, 127.5, 127.1, 126.8, 126.4, 121.0, 119.7, 26.3, 25.8,
13.7, 9.9; mmax/cm−1 1674, 1633, 1609, 1578; m/z (APCI)(%) 379
(100) [M + H+]; C20H18N4O2S requires C 63.46; H 4.79; N 14.82,
found C 63.16; H 4.73; N 14.58%.
(+)(-)-2-(1-Methyl-4-phenyl)but-3-enyl-3,3ꢀ-bisquinazoline-4,4ꢀ-
one 14. (Prepared as a 1 : 1 mixture of diastereoisomers and
not separated.)
A mixture of 2H-3,1-benzoxazin-4-one (0.76 g, 5.160 mmol),
(+)(−) 3-amino-2-(1-methyl-4-phenyl-but-3-enyl)quinazolin-4-
one (1.050 g, 3.44 mmol) and p-toluenesulfonic acid (0.026 g,
0.13 mmol) was heated at reflux in toluene for 24 hours. After
cooling, the toluene was evaporated in vacuo and the solid
obtained was crystallised from toluene–petroleum ether to give
pale yellow crystals; mp 144 ◦C (0.70 g, 47%); dH (400 MHz,
CDCl3, 1 : 1 mixture of isomers referred to as A and B, many
signals overlapped assigned as m) 8.28 (1H, d, J = 7.4 Hz, H8A),
8.26 (1H, d, J = 7.4 Hz, H8B), 8.20 (1H, d, J = 7.5 Hz, H8ꢀA),
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 3 4 – 1 1 3 9
1 1 3 7

View Article Online
8.16 (d, 1H, J = 7.5 Hz, H8ꢀB), 7.95 (1H, s, H2ꢀ), 7.74 (8H,
diethyl ether (100 ml) under nitrogen was added dropwise a
solution of (S)-acetoxyphenylacetyl chloride (8.5 g, 0.05 mol)15
as a solution in diethyl ether (50 ml) over 20 minutes at ice–
water temperature then stirred as it attained room temperature
over 3 h. After this time the reaction mixture was added to
a separating funnel and washed with hydrochloric acid (3 M,
3 × 100 ml), sodium hydrogen carbonate (saturated, aqueous,
2 × 50 ml), brine (100 ml) dried (MgSO4) and evaporated to
dryness. Crystallisation of the residue from light petroleum gave
2-(2-acetoxy-2-phenylacetylamino)benzoic acid methyl ester as
m), 7.70 (1H, s, H2ꢀ), 7.50 (5H, m), 7.14 (9H, m), 6.38 (1H, d,
=
=
J = 15.8, PhCH C), 6.28 (1H, d, J = 15.8, PhCH C), 6.04
=
(2H, m, C CHCH2), 2.84 (1H, m), 2.69 (3H, m), 2.48 (2H, m,
HC(CH3), 1.32 (3H, d, J = 6.4, CH3), 1.28 (3H, d, J = 6.4, CH3);
dC (100 MHz, CDCl3) 159.4, 159.3, 159.2, 159.1, 158.5, 158.4,
147.17, 147.12, 146.6, 146.6 (2 peaks), 145.71, 145.5, 137.1,
136.4, 135.3, 135.3, 132.9, 132.17, 128.7, 128.4, 128.2, 128.1,
128.0, 127.99, 127.94, 127.6, 127.5, 127.37, 127.32, 127.3, 127.1,
127.0, 126.8, 125.9, 125.8, 125.7, 125.0, 122.0, 120.5, 120.4, 40.5,
38.87, 37.03, 35.65, 28.19, 21.73, 20.56, 18.61, 18.49; mmax/cm−1
2920, 1693, 1603, 1568, 1462, 1377, 1266, 768, 741, 692; m/z
(EI) (%) 434.5 (38)(M+) 211 (82); HRMS C27H22N4O2 requires
435.1816, found 435.1815.
◦
white crystals; mp 91 C (11.3 g, 69%). dH (400 MHz, CDCl3)
11.85 (1H, br, NH), 8.63 (1H, d, J = 7.8 Hz, H3), 7.99 (1H,
dd, J = 8.0, 1.5 Hz, H6), 7.50 (3H, m, 2 × PhH + H4), 7.30
(3H, m, 3 × PhH), 7.05 (1H, dd, J = 8.0, 7.1 Hz, H5), 6.33
(1H, s, CH(O)Ph), 3.88 (3H, s, OCH3), 2.23 (3H, s, C(O)CH3);
dC (100 MHz, CDCl3) 169.9, 168.9, 167.9, 141.1, 135.8, 135.0,
131.2, 129.3, 127.8, 126.3, 123.5, 120.8, 115.9, 76.3, 52.7, 21.3;
mmax/cm−1 3263, 1751, 1701, 1662 cm−1; m/z(APCI)(%) 328.3
(100%)(M + H+) 117(58) 107(13); C18H17NO5 requires C 66.05;
H 5.23; N 4.28, found C 66.25; H 5.28; N 4.34%. To a solution
of 2-(2-acetoxy-2-phenylacetlamino)benzoic acid methyl ester
(2.36 g, 7 mmol) in ethanol (20 ml) was added hydrazine
monohydrate (2 m1, 2.06 g, 0.04 mol) and heated at reflux for 5 h.
The reaction mixture was concentrated under reduced pressure
and the residue crystallised from ethanol to give 17 as colourless
crystals mp 139 ◦C (1.19 g, 64%). dH (400 MHz, CDCl3) 8.19
(1H, d, J = 8.0 Hz, H5), 7.78 (2H, m, H6,8), 7.48 (1H, m, H7),
(S)-2-(1-Hydroxy-2-methyl)propyl)-3-aminoquinazolin-4-one
15. This was prepared by the procedure of Atinson11 from
(S)-2-acetoxy-3-methylbutanoic acid, prepared by the following
adaptation to the original procedure:
To a stirred suspension of L-valine (4 g, 34 mmol) in acetic
acid (160 ml) was added portionwise and with stirring over
1 hour sodium nitrite (9.4 g, 136 mmol). During the addition
the reaction vessel was cooled in a water bath which was kept
between 0 and 5 ◦C by the intermittant addition of ice, not being
allowed to exceed these limits. After the addition was complete
the bath was removed and the mixture allowed to attain ambient
temperature while it was stirred for a further 1 hour. The acetic
acid was then removed under vaccum, the residue disolved in
warm water (30 ml), extracted into ether (3 × 30 ml), washed
with brine (30 ml), dried (MgSO4) and evaporated to give (S)-2-
acetoxy-3-methylbutanoic acid (3.79 g, 69.7%) as a pale oil which
was used in subsequent steps without further purification.
=
7.37 (5H, m, PhH5), 6.01 (1H, d, J = 5.5 Hz, N C–CH(O)Ph),
5.22 (1H, d, J = 5.5 Hz, OH, exchanges D2O), 4.46 (2H, br s,
NH ); dC (100 MHz, CDCl3) 169.9, 160.7, 156.0, 144.7, 139.0,
2
133.7, 127.8, 127.4, 126.5, 126.2, 125.7, 119.1, 70.3; mmax/cm−1
3357, 3101, 1673, 1588; m/z(APCI)(%) 268.2 (100)(M + H+)
250.0 (77), 233.1 (11), 106.8 (42); C15H13N3O2 requires C 62.40;
H 4.90; N 15.72, found C 61.96; H 4.55; N 15.77%.
4-Isopropyl-3-oxo-1,9a,10-triazaanthracen-9-one 16‡.
A
solution of 3-amino-2-[(S)-1-hydroxy-2-methylpropyl]-quinazo-
lin-(3H)-4-one (0.354 g, 1.5 mmol) in triethyl orthoformate
(8 ml) was heated at reflux for five hours. After that period
the mixture was allowed to reach room temperature, and then
the solvent was removed under reduced pressure. The red residue
was crystallised from ethanol to give white crystals; mp 230 ◦C
(0.248 g, 67.5%). dH (400 MHz, CDCl3) 8.30 (1H, dd, J = 8.1,
1.3 Hz, H8), 7.70 (1H, dd, J = 8.1, 7.0 Hz, H7), 7.60 (1H, d, J =
7.7 Hz, H5), 7.50 (1H, dd, J = 7.7, 7.0 Hz, H6), 7.24 (1H, s, H2),
4.95 (1H, d, J = 4.2 Hz), 2.55 (1H, m), 1.05 (3H, d, J = 6.9 Hz,
CH3), 0.95 (3H, d, J = 6.98 Hz, CH3); dC (100 MHz, CDCl3)
155.4, 45.0, 144.9, 143.3, 133.6, 126.8, 126.5, 126.4, 121.2, 77.6,
31.9, 17.5, 15.3; mmax/cm−11699, 1654, 1598, 1463 cm−1; m/z
(APCI)(%) 244.1 [M + H+]; C13H13N3O2 HRMS requires MH+
244.1081, found: MH+244.1084; C13H13N3O2 requires C 64.12;
H 5.39; N 17.27, found C 63.83; H 5.38; N 17.12%.
4-Phenyl-3-oxo-1,9a,10-triazaanthracen-9-one 18. A solu-
tion of 3-amino-2-[1-hydroxybenzyl]-quinazolin-(3H)-4-one
(0.5 g, 1.92 mmol) in triethyl orthoformate (3 ml) was heated
at reflux for 24 hours. After that period the mixture was
allowed to reach room temperature and then the solvent was
removed under reduced pressure. The residue was crystallised
from dichloromethane–methanol to give pale orange crystals;
mp 228 ◦C (0.307 g, 58%); dH (250 MHz, CDCl3) 8.35 (1H, dd,
J = 8.2, 1.2 Hz H8), 7.70 (1H, ddd, J = 8.2, 7.0 Hz, H7), 7.60
(1H, d, J = 8.2 Hz, H5), 7.48 (1H, ddd, J = 8.2, 7.0, 1.2 Hz, H6),
7.35 (5H, m, 5 × PhH), 7.28 (1H, s, H2), 6.15 (1H, s, CH-4);
dC (100 MHz, CDCl3) 155.4, 145.1, 144.5, 142.7, 134.3, 133.7,
128.9, 128.2, 126.8, 126.7, 126.6, 125.9, 121.4, 74.2; mmax/cm−1
1708, 1648, 1600, 1463, 1368, 1172; m/z (APCI)(%) 278.0 (100%)
[M + H+]; HRMS C16H11N3O2 requires 278.0924 found MH+
278.0926; [a]2D5 = −127 (c = 1.0, CH2Cl2).
3-Oxo-1,9a,10-triazaanthracen-9-one 20. A solution of 3-
amino-2-(2-hydroxymethyl)-quinazolin-(3H)-4-one (0.510 g,
2.66 mmol) in triethyl orthoformate (6 ml) was heated at reflux
for 24 hours. After that period the mixture was allowed to reach
room temperature and then the solvent was removed under
reduced pressure. The residue was crystallised from methanol to
give pale yellow crystals; mp 200 ◦C (0.460 g, 85%). dH (400 MHz,
CDCl3) 8.30 (1H, dd, J = 8.3, 1.2 Hz, H8), 7.70 (1H, ddd, J =
8.3, 7.0, 1.2 Hz, H7), 7.60 (1H, d, J = 7.7 Hz H5), 7.46 (1H, ddd,
J = 7.7, 7.0, 1.2 Hz H6), 7.24 (1H, s, H2), 5.05 (2H, s, CH2-4); dC
(100 MHz) 156.6, 146.4, 146.3, 142.1, 135.2, 128.2, 128.0, 127.6,
123.0, 63.4; mmax/cm−1 1703, 1649, 1605,1360; m/z (APCI)(%)
202.1 (100%) [M + H+]; C10H7N3O2 requires C, 59.70; H, 3.51;
N, 20.89, found C 59.55; H 3.51; N 20.77%.
4-Isopropyl-3-oxo-1,9a,10-triazaanthracen-9-one 16 (via ben-
zoxazinones). A mixture of 3-amino-2-[(S)-1-hydroxy-2-methyl-
propyl]quinazolin-4-(3H)-one (1 g, 4.3 mmol), 4H-3,1-
benzoxazin-4-one (0.8 g, 5.44 mmol) and p-toluenesulfonic acid
(0.051 g, 0.272 mmol) was heated at reflux in toluene (30 ml) for
24 hours. After this time the reaction mixture was cooled to room
temperature then the pale brown precipitate recrystallised from
methanol to give white crystals (0.512 g, 49% yield), properties
as above.
3-Amino-2-(phenylhydroxymethyl)quinazolin-4-one 17. To a
solution of methyl anthranilate (20 ml, 23.4 g, 0.15 mol) in
‡ Crystallographic data for 16: C13H13N3O2, crystal system monoclinic;
˚
space group, P21/c; a = 9.891(5), b = 10.066(5), c = 11.732(c) A; a =
90, b = 96.743(5), c = 90◦; Z = 4; T = 150 K; l = 0.097 mm−1; l =
˚
0.71069 A (MoKa); F(000) 512; 3.85 < h < 30.02; 8656 reflections, 3323
4-Isopropyl-2-methyl-3-oxo-1,9a,10-triazaanthracen-9-one 21.
solution of 3-amino-2-[(S)-1-hydroxy-2-methylpropyl]-
unique [R(int) = 0.0626]; R1 = 0.0584, wR2 = 0.1410 [I > 2r(I)]; R1 =
0.0974, wR2 = 0.1609 (all data). CCDC reference numbers 259058. See
http://www.rsc.org/suppdata/ob/b4/b419108k/ for crystallographic
data in .cif or other electronic format.
A
quinazolin-(3H)-4-one (0.200 g, 0.85 mmol) in triethyl orthoac-
etate (3 ml) was heated at reflux for 24 hours. After that period
1 1 3 8
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 3 4 – 1 1 3 9

View Article Online
the mixture was allowed to reach room temperature and then the
solvent was removed under reduced pressure. The residue was
crystallised from ethyl acetate–petroleum ether to give white
(3H, d, J = 6.9 Hz, CH3); dC (62.5 MHz, CDCl3) 157.0, 154.2,
146.3, 145.2, 134.8, 132.6, 129.6, 128.9, 128.2, 128.0, 127.7,
127.6, 122.6, 79.9, 32.7, 19.3, 17.2; mmax 1696, 1605, 1465, 1362,
1318, 1176 cm−1; m/z (APCI)(%) 320 (100%) [M + H+]; HRMS
C19H17N3O2 requires 320.1394, found MH+ 320.1389.
◦
crystals mp 147 C (0.170 g, 77%). dH (250 MHz, CDCl3) 8.30
(1H, dd, J = 8.1, 1.2 Hz, H8), 7.68 (1H, ddd, J = 8.1, 7.0, 1.0 Hz,
H7), 7.60 (1H, dd J = 8.1, 1.0 Hz, H5), 7.50 (1H, ddd, J = 8.1,
i
=
7.0, 1.2 Hz, H6), 4.90 (2H, d, J = 4.8 Hz, N C–CH(O) Pr), 2.48
Acknowledgements
(1H, m, CHMe2), 2.2 (1H, s, Me-2), 1.05 (3H, d, J = 6.9 Hz,
CH3), 0.95 (3H, d, J = 6.8 Hz, CH3); dC (100 MHz, CDCl3)
155.5, 145.1, 143.2, 133.4, 126.7, 126.2, 126.1, 125.5, 121.7, 78.0,
31.7, 18.10, 17.6, 15.5; mmax/cm−1 1688, 1654, 1606, 1248, 1028;
m/z (EI)(%) 257.2 (18)(M+), 174.1 (80), 119.1 (60); C14H15N3O2
requires C 65.4; H 5.9; N 16.3, found C 65.1; H 5.9; N 16.1%;
[a]2D5 = −51.5 (c = 1.0, CHCl3).
We thank Robert L. Jenkins and Robin Hicks (University of
Cardiff) for Mass Spectrometry, Dr. R. S. Atkinson for the
sample of 5, the EPSRC National Mass Spectrometry Service
Centre, Chemistry Department, University of Wales, Swansea
for HRMS, and the EPSRC for support (GR/R54408/01).
4-Isopropyl-2-ethyl-3-oxo-1,9a,10-triazaanthracen-9-ones 22.
solution of 3-amino-2-[(S)-1-hydroxy-2-methylpropyl]-
References
A
1 M. P. Coogan, D. E. Hibbs and E. Smart, Chem. Commun., 1999,
1991; J. A. Platts and M. P. Coogan, J. Chem. Soc., Perkin Trans. 2,
2000, 1075; M. P. Coogan and S. C. Passey, J. Chem. Soc., Perkin
Trans. 2, 2000, 2060.
2 L. I. Giannola, G. Giammona, B. Carlisi and S. Palazzo, J. Hetero-
cycl. Chem., 1981, 18, 1557.
3 R. S. Atkinson, M. P. Coogan and C. L. Cornell, J. Chem. Soc.,
Chem Commun., 1993, 1215; R. S. Atkinson, M. P. Coogan and C. L.
Cornell, J. Chem. Soc., Perkin Trans. 1, 1996, 157.
4 M Gal, E Tihanyi and P Dvortsak, Heterocycles, 1984, 1985.
5 R. S. Atkinson, E. Barker, P. J. Edwards and G. A. Thomson, J. Chem.
Soc., Perkin Trans. 1, 1996, 1047.
6 T. George, D. V. Metha and R. Tahilramani, Ind. J. Chem., 1971,
755.
7 B. J. Kelly, Ph.D. Thesis, Leicester University, 1989.
8 D. T. Zentmyer and E. C. Wagner, J. Org. Chem., 1949, 14, 967.
9 O. Meth-Cohn and D. G. Hawkins, J. Chem. Soc., Perkin Trans. 1,
1983, 2077.
quinazolin-(3H)-4-one (0.374 g, 1.60 mmol) in triethyl ortho-
propionate (2 ml) was heated at reflux for 36 hours. The excess
of orthoester was then removed by vacuum distillation and the
brown residue crystallised from ethyl acetate–petroleum ether to
give white crystals; mp 82 ◦C (0.230 g, yield 54%). dH (400 MHz,
CDCl3) 8.30 (1H, dd, J = 8.4, 1.0 Hz H8) 7.68 (1H, dd, J = 8.4,
7.0 Hz H7) 7.58 (1H, d, J = 8.1 H5) 7.42 (ddd, 1H, J = 8.1, 7.0,
1.0 H6) 4.90 (1H, d, J = 4.6, CH-4), 2.51 (2H, q, J = 7.6 Hz, 2-
CH2), 2.44 (1H, m, CHMe2), 1.24 (3H, t, J = 7.6 Hz, CH2CH3),
1.06 (3H, d, J = 6.9 Hz, CH3), 0.97 (3H, d, J = 6.9 Hz, CH3); dC
(62.5 MHz, CDCl3) 159.6, 16.5, 146.1, 144.3, 134.3, 127.6, 127.2,
127.1, 122.2, 88.3, 32.7, 26.6, 18.6, 16.5, 10.4; mmax/cm−1 1692,
1654, 1377, 1328, 1149, 1090, 1062; m/z(APCI)(%) 272 (100%),
[M + H+]; C15H17N3O2 requires C 66.40; H 6.32; N 15.49, found
C 66.21; H 6.32; N 15.40%.
10 W. Reid and J. Valentin, Chem. Ber., 1968, 101, 2106.
11 A. G. Al-Sehemi, R. S. Atkinson, J. Fawcett and D. R. Russell,
J. Chem. Soc., Perkin Trans. 1, 2000, 4413.
4-Isopropyl-2-phenyl-3-oxo-1,9a,10-triazaanthracen-9-one 23.
solution of 3-amino-2-[(S)-1-hydroxy-2-methylpropyl]-
A
quinazolin-(3H)-4-one (0.400 g, 1.70 mmol) in trimethyl or-
thobenzoate (3 ml)was heated at reflux for 36 hours. The excess
of orthoester was then removed by vacuum distillation and the
brown residue was crystallised from diethyl ether to give pale
brown crystals mp 240 ◦C (0.324 g, 60%). dH (250 MHz, CDCl3)
8.38 (1H, dd, J = 8.0, 1.2 Hz H8), 8.12 (dd, 2H, J = 8.0, 1.6,
H7), 7.66 (2H, m), 7.45 (m, 4H), 5.06 (1H, d, J = 5.2, CH-4),
2.55 (1H, m, CHMe2), 1.20 (3H, d, J = 6.9 Hz, CH3), 1.05
12 M. Uskokovic, J. Iacobelli, V. Toome and W. Wenner, J. Org. Chem,
1964, 568.
13 A. G. Al-Sehemi, R. S. Atkinson and J. Fawcett, J. Chem. Soc., Perkin
Trans. 1, 2002, 257.
14 H. K. Gakhar, S. C. Gupta and N. Kumar, Ind. J. Chem. B, 1981, 20,
14.
15 S. W. Garry and D. G. Neilson, J. Chem. Soc., Perkin Trans. 1, 1987,
601.
16 A. A. F. Wasfy, Ind. J. Chem., Sect. B, 2003, 12, 3102.
O r g . B i o m o l . C h e m . , 2 0 0 5 , 3 , 1 1 3 4 – 1 1 3 9
1 1 3 9