T. A. Alanine et al. / Tetrahedron Letters xxx (2016) xxx–xxx
3
X
Z
NH2
NH
NH2
NH
NH2
NH2
Cl
O
X
Z
Cl
Cl
Cl
Cl
OH-
X
Z
Cl
Cl
OH
-HCl
O
O
Favoured when
Favoured when
X = EDG or Z = N X = EWG and Z = CH
3
3'
Scheme 1. Proposed mechanism for the Bargellini reaction with unsymmetrical aromatic 1,2-diamines.
represents 100% of isomer 30. These values were then plotted
against the Hammett constants12 rp for the aryl substituents.
A good correlation of the isomeric ratio with the Hammett con-
stants was observed (R2 = 0.84), showing an influence of the elec-
tronics of the ring substituents on the product ratio.13 In
particular, electron-donating substituents favoured the formation
of an excess of the 7-substituted isomer (3) and electron-with-
drawing groups favoured an excess of the 6-substituted isomer
(30). The distribution can be rationalised mechanistically
(Scheme 1), as the initial bond formed involves nucleophilic attack
of the amine group on the carbon atom distal to the chlorine atoms
of the epoxide.14 The resultant intermediate collapses to an acid
chloride moiety which can be intercepted by the less nucleophilic
aromatic amine group to give the observed products.
In summary, the synthesis of 3,3-substituted dihydroquinoxali-
nones from aromatic 1,2-diamines using the Bargellini reaction has
been revisited and the scope expanded. The reaction proceeds
under mild conditions and tolerates a variety of trichloromethyl-
carbinol electrophiles to give useful quaternary 3,3-disubstituted
dihydroquinoxalinones. The mild conditions and generally good
yields of the process highlights the Bargellini process as an efficient
means to access this important class of heterocyclic substrates. In
the case of unsymmetrically substituted arenes, a mixture of
regioisomers was obtained. The isomer ratio can be rationalised
in terms of the electronic effects of the aryl substituents. Pyridine
substrates are also tolerated, and these showed excellent intrinsic
selectivity for the dihydro[3,2-b]pyridopyrazin-2-one isomer.
Efforts to prepare stereodefined trichloromethylcarbinols are cur-
rently underway in order to access enantiomerically pure
dihydroquinoxalinones.
Hagiwara, Y.; Oki, K.; Matsuda, M.; Mori, T. EP2327699, 2011; (e) Billhardt, U.-
M.; Roesner, M.; Riess, G.; Winkler, I.; Bender, R. US6369057, 2002
3. (a) Combs, A. P.; Maduskuie, Jr., T. P.; Falahatpisheh, N. WO2015/164480, 2015;
(b) Matsuda, M.; Mori, T.; Nagatsuka, M.; Kobayashi, S.; Takaoka, S.; Kato, M.;
Takai, M.; Matsuyama, T.; Kurose, T.; Hagiwara, Y. WO2009035068A1, 2009; (c)
7. In Ref. 6, a note is made that regioisomeric mixtures are obtained, but no
details are given on the relative amounts.
9. In our hands, efforts to perform the Bargellini reaction directly from chloroform
and ketones were much less successful, leading to complex reaction mixtures,
incomplete conversion and low yields.
10. General procedure for the synthesis of 3a–3t: Representative procedure for 3h:
50% aqueous sodium hydroxide (0.42 mL, 5 equiv) was added dropwise to a
stirred mixture of o-phenylene diamine 1a (CAUTION: toxic) (200 mg, 1 equiv),
4-(trichloromethyl)tetrahydro-2H-pyran-4-ol (812 mg, 2 equiv) and N-benzyl-
N,N-diethylethanaminium chloride (42 mg, 0.1 equiv) in CH2Cl2 (0.1 M) cooled
to 0 °C over a period of 10–20 s under nitrogen. The resulting mixture was
stirred at 0 °C and left to warm to room temperature with stirring over 18 h.
The reaction was monitored by analytical HPLC–MS. When complete, the
reaction mixture was diluted with water (10 mL) until any solid had dissolved
and the layers were separated. The aqueous layer was extracted with CH2Cl2
(3 ꢁ 20 mL). The combined organic layers were dried (MgSO4), filtered and
evaporated to afford the crude product which was purified by flash silica
chromatography, elution gradient 0–50% EtOAc in heptane. The resulting solid
was filtered through a Buchner funnel, rinsed with MTBE (2 ꢁ 10 mL), and
collected to afford 10,2,3,40,5,6-hexahydro-30H-spiro[pyran-4,20-quinoxalin]-30-
one (310 mg, 77%) as a pale yellow powder. Rf = 0.33 (1:1 EtOAc/heptane). 1H
NMR (500 MHz, d6-DMSO): dH 10.21 (1H, s), 6.93–6.33 (1H, m), 6.78 (1H, td,
J = 1.4, 7.6 Hz), 6.73 (1H, dd, J = 1.3, 7.7 Hz), 6.62 (1H, td, J = 1.3, 7.6 Hz), 6.19
(1H, s), 3.72 (4H, dddd, J = 5.0, 10.6, 16.6, 21.6 Hz), 1.90 (2H, ddd, J = 4.5, 9.1,
13.6 Hz), 1.43 (2H, dt, J = 3.5, 13.5 Hz). 13C NMR (126 MHz, d6-DMSO): dc 168.9,
133.0, 126.0, 122.7, 118.1, 114.4, 114.3, 62.2 (2 ꢁ C), 53.3, 32.2 (2 ꢁ C). IR
(cmꢀ1, CH2Cl2): 3340, 2933, 1671, 1503, 1421, 1370, 1311, 1242, 1160, 1101.
HRMS (TOF ES+) Calcd for C12H15N2O2 [M+H]+: 219.1134, found 219.1139. For
substrates 3r and 3s, the aqueous layer was adjusted to pH 7 with saturated
aqueous ammonium chloride (10 mL) and the extraction performed with
EtOAc (3 ꢁ 20 mL).
Acknowledgments
The authors would like to acknowledge Dr. Eva Lenz and Dr.
Paul Davey for NMR and MS support, as well as AstraZeneca IMED
for funding.
Supplementary data
Supplementary data (NMR resonances and accurate masses)
associated with this article can be found, in the online version, at
11. In the case of 3p, both isomers were successfully separated by column
chromatography.
13. Similar correlations were obtained with Hammett constants r+p and rpꢀ
(R2 = 0.83 and 0.84, respectively).
References and notes