A simple approach for the synthesis of 7,8-diaza[5]helicene

Article abstract of DOI:10.1055/s-2008-1032021

An oxidative route for the formation of 7,8-diaza[5]helicene and the corresponding N-oxides in high yields was developed. A mechanism of formation and possible interconversion of these products is put forward. Reduction of the N-oxides provided the title compound in quantitative yield. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2008-1032021

PAPER
413
A Simple Approach for the Synthesis of 7,8-Diaza[5]helicene
S
ynthesis of 7,8
u
-Diaza[5]helicene lio Caronna,*a Francesca Fontana,a Andrea Mele,b,c Isabella Natali Sora,a Walter Panzeri,b Luca Viganòa
a
Dipartimento di Ingegneria Industriale, Università di Bergamo, viale Marconi 5, 24044 Dalmine, BG, Italy
E-mail: tullio.caronna@unibg.it; fax +39(035)562779
b
c
Dipartimento di Chimica, Ingegneria Chimica e Materiali G. Natta, Politecnico di Milano, via L. Mancinelli 7, 20131 Milano, Italy
CNR – Istituto di Chimica del Riconoscimento Molecolare, via L. Mancinelli 7, 20131 Milano, Italy
Received 12 October 2007; revised 30 October 2007
reported by Corbett and Holt to yield only traces of diaz-
Abstract: An oxidative route for the formation of 7,8-diaza[5]heli-
cene and the corresponding N-oxides in high yields was developed.
A mechanism of formation and possible interconversion of these
products is put forward. Reduction of the N-oxides provided the title
compound in quantitative yield.
ene 2.8
In the present communication, the use of other reagents
capable of oxidizing diamine 1 in high yields is explored
and discussed. The proposed approach starts from readily
available reagents. A convenient route for the preparation
of either 7,8-diaza[5]helicene or intermediate products
that can be converted into the target helicene is illustrated.
Key words: diaza[5]helicenes, oxidation, amines, N-oxides, het-
erocycles
Scheme 1 shows the products detected and isolated via
flash chromatography after the oxidation of diamine 1
with m-chloroperoxybenzoic acid (MCPBA). The forma-
tion of side products was assessed by HPLC; however,
GC/MS analysis did not allow us to identify the impuri-
ties, probably due to their thermal instability. It turned out
that the formation of side products decreased when there
was an increase in the dilution and the dropwise addition
rate of the oxidizing agent. The identification of the N-ox-
ides 3 and 4 was achieved by electrospray ionization mass
spectrometry (ESI-MS) and NMR spectroscopy.
In the framework of our research on the azahelicene sys-
tem, we were interested in obtaining 7,8-diaza[5]helicene
(2). The position of the two nitrogen atoms, adjacent and
located on the central ring, may arouse interest both from
a theoretical standpoint and for practical purposes, with
emphasis on the following points: (i) determination of the
energy barrier for the interconversion between the P- and
M- enantiomers, (ii) the circular dichroism spectra and
(iii) the formation of metal complexes and the study of
ligand–metal energy transfer, with the purpose of obtain-
ing tunable light emission as a function of the selected
metal ion.
A certain number of syntheses of 7,8-diaza[5]helicene are
reported in the literature. Some early papers deal with the
reduction of 2-nitronaphthalene with different reducing
agents.1 Later (1965), phosphine was used as the reducing
agent.2 However, at present, 2-nitronaphthalene is expen-
sive and not available on a large scale, possibly because it
is toxic and commercially uninteresting. The reduction of
2,2¢-dinitro-1,1¢-binaphthyl with lithium aluminum hy-
dride has also been reported,3 but the starting material is
nowadays not readily available. The synthesis of ben-
zo[c]cinnoline, and the corresponding N-oxide, was car-
H2N
H2N
1
MCPBA
4
5
3
2
6
O
7
8
N
N
N
N
1
N
N
+
+
14
13
12
ried out starting from 2,2¢-dinitro-1,1¢-biphenyl.4
A
O
O
coupling method starts from 2,2¢-azonaphthalene and alu-
minum trichloride,5 but once again the starting material is
not so easily achievable. The oxidative approach, namely
the oxidation of diamino derivatives, was seldom consid-
ered promising. Some papers have reported the oxidation
of amines to form azoxy or N,N¢-dioxide derivatives, ei-
ther using hydrogen peroxide and polyoxometalates,6 or
with the use of aliphatic peracids.7 The oxidation of 2,2¢-
diamino-1,1¢-binaphthyl (1) with sodium perborate was
9
11
10
2
4
3
Scheme 1 Oxidation of 2,2¢-diamino-1,1¢-binaphthyl (1) with
MCPBA
The oxidation of amino derivatives has been widely stud-
ied for a long time,9 and it is well known that different
products are formed depending upon the oxidant and the
experimental conditions. One of the steps of the reaction
mechanism generally accepted for this oxidation consists
of the transformation of the amino group into a nitroso de-
rivative, possibly via a hydroxylamino derivative.4,6 The
SYNTHESIS 2008, No. 3, pp 0413–0416
x
x
.
x
x
.2
0
0
7
Advanced online publication: 10.01.2008
DOI: 10.1055/s-2008-1032021; Art ID: Z23907SS
© Georg Thieme Verlag Stuttgart · New York

414
T. Caronna et al.
PAPER
dimerization of two nitroso groups to the corresponding
N,N¢-dioxide derivative has been studied in depth,6,10 and
was also observed under reductive conditions.2 The pa-
pers mentioned above also reported formation of the N,N¢-
dioxide species upon treatment of hydroxylamino deriva-
tives with hydrogen peroxide in chloroform, whilst the
azoxy derivative was isolated when water was used as the
solvent. The eventual formation of the azo functional
group may be explained by the direct condensation of a ni-
troso derivative with the starting amine.11 An interconver-
sion between the products formed was also hypothesized.
In the present work, control experiments were carried out
in order to understand if the products formed may inter-
convert. Indeed, 2, 3 and 4 were reacted, in separate
batches, with MCPBA under our experimental conditions.
It was found that 2 was partially transformed into its N-
monooxide, 10% of the N-monooxide 3 was transformed
into the corresponding N,N¢-dioxide 4, along with more
than 20% mass loss of 3, and, finally, only 10% mass loss
was observed when 4 was reacted. For a better under-
standing of the mechanism of this oxidation, two series of
reactions were carried out, the first with different amounts
of MCPBA at 50 °C and the second at three different tem-
peratures while keeping the amine-to-oxidant molar ratio
at 1:3. The two series of reactions were repeated twice.
The quantitation of the reaction products for the first and
second reaction series are reported in Figure 1 and
Figure 2, respectively. Quantitative analysis was under-
taken at three different wavelengths for 2, 3 and 4, while
the amount of diamine 1 was determined only at 254 and
350 nm due to its low extinction coefficient at 313 nm.
Figure 1 Quantitation of the oxidation products
Figure 2 Variation trends of the oxidation products with increasing
reaction temperature
of the experimental conditions, and so small, means that
its possible transformation to one or both of the other two
products will result in very little influence on their
amounts; for this reason, we will not take this factor into
consideration in further discussions. In Scheme 2, a sim-
plified reaction mechanism is drawn.
One would expect that a smaller amount of oxidant should
lead to diazene 2, while upon increasing the amount of ox-
idant, the reaction would shift toward the more oxidized
products 3 and 4. On the contrary, even at low oxidant
concentration and at low temperature, the amount of dia-
zene 2 is very low and nearly independent of an increase
in the amount of oxidant or in the temperature. The quan-
tity of diazene 2 being essentially constant, independent
Figure 2 indicates that the formation of N,N¢-dioxide 4
dominates at low temperature. This may be easily ex-
plained considering that product 4 derives from the dimer-
ization of two nitroso groups, namely two free radicals,
and the activation energy for this dimerization should be
rather low. In contrast, compound 3 is expected to arise
from an elimination reaction between one nitroso and one
H2N
H2N
ON
ON
H2N
ON
4
2
HOHN
ON
3
Scheme 2 Outline of the reaction mechanism
Synthesis 2008, No. 3, 413–416 © Thieme Stuttgart · New York

PAPER
Synthesis of 7,8-Diaza[5]helicene
415
dropwise during 1 h. The resulting mixture was stirred for an addi-
tional 1 h. After cooling, H2O (50 mL) and sat. aq NaHCO3 (50 mL)
were added and the resulting solution was extracted with CH2Cl2 (3
× 30 mL). The organic layer was dried (Na2SO4) and concentratred
under reduced pressure. The residue (0.49 g) was flash-chromato-
graphed on silica gel (EtOAc/hexane, 3:2) giving 1 (0.03 g, 7%), 2
(0.01 g, 2%), 3 (0.16 g, 34%) and 4 (0.25 g, 50%). Under these con-
ditions, other oxidation products or byproducts were not eluted
from the column.
hydroxylamino group. This reaction is slower than the
dimerization of two free radicals and, indeed, product 3
becomes the prevailing product at higher temperature. A
reasonable explanation of the trend displayed in Figure 1
stems from the following considerations: at low oxidant
concentration a partial oxidation to the nitroso–hydroxyl-
amino derivative is likely to occur. The NO and NHOH
groups may, in turn, undergo condensation leading to 3.
This reaction, however, is still slow. Hence, increasing the
amount of the oxidizing agent results in conversion of the
hydroxylamino group into a nitroso group, thus increasing
the amount of 4. In the same way, the observed low
amount of diazene 2 can be rationalized: the rate of con-
Reactions with Different Amounts of Peroxide
The molar ratios and amounts of NaHCO3 and MCPBA are reported
in Table 1. Diamine 1 (0.015 g, 0.05 mmol) was dissolved in a mix-
ture of H2O (4.5 mL) and MeCN (4.5 mL). The temperature was
raised to 50 °C. Solid NaHCO3 was added and then the soln of
densation is expected to be low enough to be detrimental MCPBA in MeCN (10 mL) was added dropwise over 30 min; the
resulting mixture was stirred further for 1 h. After cooling, sat. aq
for the direct formation of the helicene. Considering the
NaHCO3 (15 mL) was added. A soln of pyrene as internal standard
trends of the control experiments, we can conclude that
was added and the resulting solution was extracted with CH2Cl2 (3
× 10 mL). The organic layer was dried (Na2SO4) and concentrated
under reduced pressure. The residue was dissolved in MeCN and
then analyzed by HPLC at three different wavelengths (254, 313
and 350 nm).
the amount of 4 destroyed by the oxidation is counterbal-
anced by the amount of 3 transformed into 4; on the other
hand, part of 3 is directly destroyed and this explains the
trends shown in Figure 1.
Interestingly and different from that reported for the oxi-
dation of aromatic amines,6 N,N¢-dioxide 4 could be ob-
tained. By using hydrogen peroxide in the presence of
Table 1 Reactions with Different Amounts of Peroxide
Molar ratios
70% MCPBA
NaHCO3
g (mmol)
metal complexes as the oxidizing system, 4 was obtained (1/peroxide/NaHCO3) g (mmol)
only in chloroform solution. A possible explanation may
1:1.5:1.5
1:3:3
0.019 (0.08)
0.039 (0.16)
0.061 (0.25)
0.103 (0.40)
0.123 (0.50)
0.006 (0.08)
0.013 (0.16)
0.021 (0.25)
0.034 (0.40)
0.042 (0.50)
be based on the fact that in the substrate described in the
present work the two naphthyl groups are not coplanar for
steric reasons. This allows oxidation of the hydroxylami-
no group to the corresponding nitroso group and easy
cross-coupling of the nitroso radicals leading to the for-
mation of 4.
1:5:5
1:8:8
1:10:10
2,2¢-Diamino-1,1¢-binaphthyl (1) and m-chloroperoxybenzoic acid
(70% w/w) (MCPBA) are commercial products. HPLC analyses
were performed on an Agilent 1200n series instrument with an
Agilent ZORBAX Eclipse XDB-C18 5 mm, 4.6 × 150 mm column,
using MeCN as solvent at a flow rate of 0.5 mL/min. The GC/MS
instrument was an Agilent 6850 Network GC system, with an Agi-
lent 19091 S-433E column, 325 °C max, flow 1 mL/min, tempera-
ture programming from 90 to 290 °C, 25 °C/min after the first
minute, using a 5973 Mass Select Detector. The ESI-MS apparatus
was a Bruker Esquire 3000+ instrument with an electrospray source
and a quadrupole ion trap detector. The samples were dissolved in
50:50 MeOH–H2O (containing 1% formic acid) and infused into the
ESI source via a microsyringe pump at a rate of 4 mL/min. NMR
spectra were recorded either on a Bruker ARX 400 or Bruker
Avance 500 spectrometer operating at proton resonance frequencies
of 400 and 500 MHz (100 and 125 MHz for 13C NMR, respectively)
using CDCl3 as solvent and TMS as internal standard. Ultraviolet/
visible absorption spectra were recorded on a Thermo Nicolet Evo-
lution UV/vis-500 spectrophotometer, with MeCN as solvent and
Vision 32 software, using a 1-cm path length cell. FT-IR spectra
were recorded on a Thermo Avatar 370 instrument in ATR reflec-
tion mode using a zinc selenide crystal. All spectra were recorded at
r.t.
Reactions at Different Temperatures
The conditions were the same as those reported above, using a mo-
lar ratio of 1:3:3 (1/peroxide/NaHCO3). Workup was the same as
above. The temperatures chosen were –2 °C, 20 °C and 50 °C.
Control Experiments
Compound 2 or 3 or 4 (0.048 mmol) and NaHCO3 (0.048 mmol)
were dissolved in a mixture of H2O (4.5 mL) and MeCN (4.5 mL).
A large amount of MCPBA (0.385 mmol, 8:1 ratio with the selected
helicene) in MeCN (10 mL) was added dropwise to the solution dur-
ing 30 min, and the mixture was left at 50 °C for a further 1 h. After
cooling, sat. aq NaHCO3 (15 mL) was added. A soln of pyrene as
internal standard was added and the resulting solution was extracted
with CH2Cl2 (3 × 10 mL). The organic layer was dried (Na2SO4) and
concentrated under reduced pressure. The residue was dissolved in
MeCN and then analyzed by HPLC at three different wavelengths
(254, 313 and 350 nm).
1H NMR shifts and coupling constants for compounds 2, 3 and 4 are
reported in Table 2 (atom numbering is given in Scheme 1).
7,8-Diaza[5]helicene (2)
Mp 268–269 °C (Lit.2 269–270 °C).
IR: 3058, 2960, 2922, 2850, 1647, 1540, 1505, 1455, 1415, 1337,
1257, 1239, 1162, 1092, 1079, 1035, 957, 878, 845, 821, 810, 783,
775, 766, 748, 651 cm–1.
13C NMR (CDCl3): d = 146.65, 134.19, 130.68, 129.32, 129.20,
128.16, 127.96, 126.47, 125.37, 119.90.
Preparative Reaction
2,2¢-Diamino-1,1¢-binaphthyl (1) (0.45 g, 1.6 mmol) was dissolved
in a mixture of H2O (100 mL) and MeCN (100 mL) and the temper-
ature was brought to 50 °C. When this temperature was reached,
first NaHCO3 (0.67 g, 7.9 mmol) was added and then a solution of
70% MCPBA (1.95 g, 7.9 mmol) in MeCN (200 mL) was added
Synthesis 2008, No. 3, 413–416 © Thieme Stuttgart · New York

416
T. Caronna et al.
PAPER
Table 2 1H NMR Data for Compounds 2, 3 and 4: Chemical Shift (d, ppm), Multiplicity, Coupling Constants (Hz)
2
3
4
H1
H2
8.81, d, J = 8.5
8.67, d, J = 8.5
8.38, d, J = 8.6
7.43, ddd, J = 8.5, 1.4, 6.9
7.47, ddd, J = 8.5, 1.4, 6.9
7.74, ddd, J = 8.1, 1.2, 6.9
8.05, dd, J = 8.1, 1.4
8.15, d, J = 9.2
7.44, ddd, J = 8.6, 1.4, 6.9
H3
7.71, ddd, J = 8.1, 1.2, 6.9
7.69, ddd, J = 8.1, 1.2, 6.9
H4
8.03, dd, J = 8.1, 1.4
8.03, dd, J = 8.1, 1.4
H5
8.16, d, J = 8.9
8.19, d, J = 9.2
H6
8.54, d, J = 8.9
8.80, d, J = 9.2
8.56, d, J = 9.2
H9
–
–
–
–
–
–
8.10,a d, J = 8.9
–
–
–
–
–
–
H10
H11
H12
H13
H14
7.95,a d, J = 8.9
7.98, dd, J = 8.1, 1.4
7.64, ddd, J = 8.1, 1.2, 6.9
7.41, ddd, J = 8.5, 1.4, 6.9
8.46, d, J = 8.5
a Assignments may be reversed.
ESI-MS: m/z = 281 [M + H]+.
Reduction of the N-Oxides
Either N-oxide 3 (0.15 g, 0.5 mmol) or N,N¢-dioxide 4 (0.16 g, 0.5
mmol) was dissolved in anhyd Et2O (20 mL) and an excess of
LiAlH4 (0.04 g, 1 mmol) was added. The mixture was heated at re-
flux (36 °C) for 5 h. After cooling, unreacted LiAlH4 was destroyed
by adding MeOH (10 mL); H2O (20 mL) was added and the ether
layer, after separation and drying (Na2SO4), was concentrated to
give 7,8-diaza[5]helicene (2) in quantitative yield.
UV/Vis (MeCN): lmax (log e) = 271 (4.24), 305 (4.32), 324s (4.08),
395 (3.35), 417 nm (3.33).
7,8-Diaza[5]helicene N-Oxide (3)
Yellow-brown solid; mp 248–251 °C [Lit.8 250–252 °C (dec)].
IR: 3062, 2923, 2853, 1613, 1595, 1511, 1448, 1424, 1369, 1335,
1261, 1237, 912, 879, 822, 807, 766, 747, 719, 679, 666, 651 cm–1.
13C NMR (CDCl3): d = 144.08, 137.22, 134.53, 133.47, 133.19,
132.33, 131.57, 130.19, 129.85, 129.15, 128.47, 128.42, 128.32,
128.13, 128.13, 126.09, 125.88, 123.40, 117.59, 115.43.
References
(1) (a) Meisenheimer, J.; Witte, K. Ber. Dtsch. Chem. Ges.
1903, 36, 4153. (b) Cumming, W. M.; Ferrier, G. S. J.
Chem. Soc. 1924, 1108.
(2) Bellaart, A. C. Tetrahedron 1965, 21, 3285.
(3) Braithwaite, P. S. W.; Holt, P. F. J. Chem. Soc. 1959, 3025.
(4) Bjørsvik, H.-R.; Gonzales, R. R.; Liguori, L. J. Org. Chem.
2004, 69, 7720.
(5) Holt, P. F.; Went, C. W. J. Chem. Soc. 1963, 4099.
(6) Sakaue, S.; Tsubakino, T.; Nishiyama, Y.; Ishii, Y. J. Org.
Chem. 1993, 58, 3633.
(7) Greenspan, F. P. Ind. Eng. Chem. 1947, 39, 847.
(8) Corbett, J. F.; Holt, P. F. J. Chem. Soc. 1961, 3695.
(9) For example, see: Bamberger, E.; Rising, A. Ber. Dtsch.
Chem. Ges. 1900, 33, 3623.
ESI-MS: m/z = 297 [M + H]+.
UV/Vis (MeCN): lmax (log e) = 252 (4.14), 281 (4.26), 330 (4.14),
350s (3.96), 417 (3.48), 442 nm (5.51).
7,8-Diaza[5]helicene N,N¢-Dioxide (4)
Creamy yellow solid; mp 230–232 °C (dec).
IR: 3082, 1610, 1592, 1577, 1507, 1462, 1443, 1427, 1396, 1363,
1325, 1263, 1168, 1144, 1123, 1070, 1048, 1018, 963, 952, 932,
889, 878, 840, 826, 811, 784, 770, 746, 657 cm–1.
13C NMR (CDCl3): d = 136.08, 133.24, 132.79, 128.75, 128.65,
128.61, 128.38, 126.47, 117.78, 115.83.
ESI-MS: m/z = 313 [M + H]+.
(10) Calcasi, M.; Tordo, P.; Gronchi, G. J. Phys. Chem. 1986, 90,
1403.
(11) Houben–Weyl Methoden der organischen Chemie, Vol. X/3;
Müller, E., Ed.; Georg Thieme: Stuttgart, 1965, and
references cited therein.
UV/Vis (MeCN): lmax (log e) = 251 (4.17), 282 (4.00), 334 (4.06),
375s nm (3.54).
Anal. Calcd for C20H12N2O2: C, 76.91; H, 3.87; N, 9.97. Found: C,
76.73; H, 3.89; N, 8.98.
Synthesis 2008, No. 3, 413–416 © Thieme Stuttgart · New York